Model profile · OpenAI

GPT-5.4 nano

Closed weightsbudget · registry tag 2026 nano

Verified but aging

Reads as verified but aging across the resolved source data. Capability fit excludes speed and cost.

Visible coverage: 63.7%
Verified coverage: 56.2%
Spread: 83.1%
Last verified: Oct 3, 2026

33%capability fit

textcodevisiondocumentsearch18 aliases42 official source links

Open compare

Data version

Current snapshot.

Published Oct 3, 2026Model list checked4 stale sources105 claim warnings9 providers · 1334 tracked models

The registry snapshot and page stamp are shown so a stale deploy is visible at a glance. View live data status.

Source-linked scores by benchmark

Each row keeps the benchmark source, source type, raw metric, and percentile inside its fair comparison set. Capability summaries exclude archived categories, speed, and cost; those rows remain below.

Verified but agingThis model currently reads as verified but aging across the resolved source data.

Chat / text36 benchmarks35.5%

Intelligence Index

AA · Chat / text · Combined

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #405 · Source label: GPT-5 nano (Minimal)

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 7
Percentile: 27.2%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `intelligenceIndex`.

27.2% percentile inside its fair comparison set

7Raw benchmark value

AA-Omniscience accuracy

AA · Chat / text · Objective

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #363 · Source label: GPT-5 nano (Minimal)

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 13.1%
Percentile: 20.1%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `omniscienceAccuracy`.

20.1% percentile inside its fair comparison set

13.1%Raw benchmark value

AA-Omniscience non-hallucination

AA · Chat / text · Objective

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #313 · Source label: GPT-5 nano (Minimal)

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 11.2%
Percentile: 31.1%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `omniscienceNonHallucination`.

31.1% percentile inside its fair comparison set

11.2%Raw benchmark value

IFBench

AA · Chat / text · Objective

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #266 · Source label: GPT-5 nano (Minimal)

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 32.5%
Percentile: 23.9%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `ifbench`.

23.9% percentile inside its fair comparison set

32.5%Raw benchmark value

Input price

AA · Chat / text · Speed / cost

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #98 · Source label: GPT-5.4 nano (Xhigh)

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: $0.2 /1M input tokens
Percentile: 72.6%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `price1mInputTokens`.

72.6% percentile inside its fair comparison set

$0.2 /1M input tokensRaw benchmark value

Output price

AA · Chat / text · Speed / cost

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #148 · Source label: GPT-5.4 nano (Xhigh)

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: $1.3 /1M output tokens
Percentile: 58.5%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `price1mOutputTokens`.

58.5% percentile inside its fair comparison set

$1.3 /1M output tokensRaw benchmark value

Output Speed

AA · Chat / text · Speed / cost

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #57 · Source label: GPT-5 nano (Minimal)

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 143.2 tokens/s
Percentile: 78.6%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

78.6% percentile inside its fair comparison set

143.2 tokens/sRaw benchmark value

Time to first token

AA · Chat / text · Speed / cost

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #245 · Source label: GPT-5 nano (High)

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 89.60s
Percentile: 7.6%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `medianTimeToFirstTokenSeconds`.

7.6% percentile inside its fair comparison set

89.60sRaw benchmark value

Time to first answer token

AA · Chat / text · Speed / cost

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #239 · Source label: GPT-5 nano (High)

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 89.60s
Percentile: 9.2%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `medianTimeToFirstAnswerTokenSeconds`.

9.2% percentile inside its fair comparison set

89.60sRaw benchmark value

Text Arena

AR · Chat / text · Human

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #141 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,401
Percentile: 62.8%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: overall. Source rank: #158. Votes: 62249. Organization: openai. License: Proprietary.

62.8% percentile inside its fair comparison set

1,401Raw benchmark valueCI 1,398 - 1,405

Text Arena · Creative Writing

AR · Chat / text · Human

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #172 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,336
Percentile: 54.3%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: creative_writing. Source rank: #194. Votes: 10368. Organization: openai. License: Proprietary.

54.3% percentile inside its fair comparison set

1,336Raw benchmark valueCI 1,329 - 1,344

Text Arena · English

AR · Chat / text · Human

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #145 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,413
Percentile: 61.7%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: english. Source rank: #161. Votes: 28808. Organization: openai. License: Proprietary.

61.7% percentile inside its fair comparison set

1,413Raw benchmark valueCI 1,408 - 1,418

Text Arena · Exclude Ties

AR · Chat / text · Human

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #137 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,389
Percentile: 63.8%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: exclude_ties. Source rank: #153. Votes: 47863. Organization: openai. License: Proprietary.

63.8% percentile inside its fair comparison set

1,389Raw benchmark valueCI 1,383 - 1,394

Text Arena · Hard Prompts

AR · Chat / text · Human

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #138 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,421
Percentile: 63.6%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: hard_prompts. Source rank: #153. Votes: 41240. Organization: openai. License: Proprietary.

63.6% percentile inside its fair comparison set

1,421Raw benchmark valueCI 1,417 - 1,426

Text Arena · Hard Prompts English

AR · Chat / text · Human

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #142 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,429
Percentile: 62.3%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: hard_prompts_english. Source rank: #157. Votes: 19866. Organization: openai. License: Proprietary.

62.3% percentile inside its fair comparison set

1,429Raw benchmark valueCI 1,424 - 1,435

Text Arena · Instruction Following

AR · Chat / text · Human

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #143 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,387
Percentile: 62.2%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: instruction_following. Source rank: #161. Votes: 21407. Organization: openai. License: Proprietary.

62.2% percentile inside its fair comparison set

1,387Raw benchmark valueCI 1,381 - 1,392

Text Arena · Longer Query

AR · Chat / text · Human

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #147 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,403
Percentile: 58.9%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: longer_query. Source rank: #166. Votes: 27656. Organization: openai. License: Proprietary.

58.9% percentile inside its fair comparison set

1,403Raw benchmark valueCI 1,397 - 1,408

Text Arena · Multi Turn

AR · Chat / text · Human

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #129 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,413
Percentile: 65.8%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: multi_turn. Source rank: #144. Votes: 11793. Organization: openai. License: Proprietary.

65.8% percentile inside its fair comparison set

1,413Raw benchmark valueCI 1,406 - 1,420

Text Arena · No Style Control

AR · Chat / text · Human

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #161 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,372
Percentile: 57.4%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: overall. Source rank: #179. Votes: 62249. Organization: openai. License: Proprietary.

57.4% percentile inside its fair comparison set

1,372Raw benchmark valueCI 1,368 - 1,376

Text Arena · Creative Writing · No Style Control

AR · Chat / text · Human

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #185 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,314
Percentile: 50.8%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: creative_writing. Source rank: #208. Votes: 10368. Organization: openai. License: Proprietary.

50.8% percentile inside its fair comparison set

1,314Raw benchmark valueCI 1,307 - 1,321

Text Arena · English · No Style Control

AR · Chat / text · Human

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #167 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,381
Percentile: 55.9%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: english. Source rank: #184. Votes: 28808. Organization: openai. License: Proprietary.

55.9% percentile inside its fair comparison set

1,381Raw benchmark valueCI 1,376 - 1,386

Text Arena · Exclude Ties · No Style Control

AR · Chat / text · Human

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #161 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,345
Percentile: 57.4%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: exclude_ties. Source rank: #178. Votes: 47863. Organization: openai. License: Proprietary.

57.4% percentile inside its fair comparison set

1,345Raw benchmark valueCI 1,340 - 1,350

Text Arena · Hard Prompts · No Style Control

AR · Chat / text · Human

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #155 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,381
Percentile: 59%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: hard_prompts. Source rank: #172. Votes: 41240. Organization: openai. License: Proprietary.

59% percentile inside its fair comparison set

1,381Raw benchmark valueCI 1,376 - 1,385

Text Arena · Hard Prompts English · No Style Control

AR · Chat / text · Human

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #158 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,387
Percentile: 58%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: hard_prompts_english. Source rank: #175. Votes: 19866. Organization: openai. License: Proprietary.

58% percentile inside its fair comparison set

1,387Raw benchmark valueCI 1,382 - 1,393

Text Arena · Instruction Following · No Style Control

AR · Chat / text · Human

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #155 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,362
Percentile: 59%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: instruction_following. Source rank: #173. Votes: 21407. Organization: openai. License: Proprietary.

59% percentile inside its fair comparison set

1,362Raw benchmark valueCI 1,356 - 1,367

Text Arena · Longer Query · No Style Control

AR · Chat / text · Human

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #161 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,366
Percentile: 54.9%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: longer_query. Source rank: #181. Votes: 27656. Organization: openai. License: Proprietary.

54.9% percentile inside its fair comparison set

1,366Raw benchmark valueCI 1,360 - 1,371

Text Arena · Multi Turn · No Style Control

AR · Chat / text · Human

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #153 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,382
Percentile: 59.4%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: multi_turn. Source rank: #170. Votes: 11793. Organization: openai. License: Proprietary.

59.4% percentile inside its fair comparison set

1,382Raw benchmark valueCI 1,375 - 1,389

Instruction following

LB · Chat / text · Objective

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #38 · Source label: gpt-5.4-nano-xhigh

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 67.2%
Percentile: 43.1%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Category: IF. Tasks scored: 4.

43.1% percentile inside its fair comparison set

67.2%Raw benchmark value

Language

LB · Chat / text · Objective

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #66 · Source label: gpt-5.4-nano-xhigh

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 62.5%
Percentile: 0%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Category: Language. Tasks scored: 3.

0% percentile inside its fair comparison set

62.5%Raw benchmark value

Paraphrase

LB · Chat / text · Objective

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #34 · Source label: gpt-5.4-nano-xhigh

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 66.6%
Percentile: 49.2%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Task: paraphrase. Category: IF.

49.2% percentile inside its fair comparison set

66.6%Raw benchmark value

Simplify

LB · Chat / text · Objective

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #38 · Source label: gpt-5.4-nano-xhigh

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 62.8%
Percentile: 43.1%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Task: simplify. Category: IF.

43.1% percentile inside its fair comparison set

62.8%Raw benchmark value

Story generation

LB · Chat / text · Objective

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #49 · Source label: gpt-5.4-nano-xhigh

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 67.7%
Percentile: 26.2%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Task: story_generation. Category: IF.

26.2% percentile inside its fair comparison set

67.7%Raw benchmark value

Summarize

LB · Chat / text · Objective

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #30 · Source label: gpt-5.4-nano-xhigh

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 71.8%
Percentile: 55.4%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Task: summarize. Category: IF.

55.4% percentile inside its fair comparison set

71.8%Raw benchmark value

Connections

LB · Chat / text · Objective

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #48 · Source label: gpt-5.4-nano-xhigh

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 94.4%
Percentile: 27.7%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Task: connections. Category: Language.

27.7% percentile inside its fair comparison set

94.4%Raw benchmark value

Plot unscrambling

LB · Chat / text · Objective

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #66 · Source label: gpt-5.4-nano-xhigh

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 39.1%
Percentile: 0%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Task: plot_unscrambling. Category: Language.

0% percentile inside its fair comparison set

39.1%Raw benchmark value

Typos

LB · Chat / text · Objective

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #66 · Source label: gpt-5.4-nano-xhigh

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 54%
Percentile: 0%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Task: typos. Category: Language.

0% percentile inside its fair comparison set

54%Raw benchmark value

Coding28 benchmarks38.5%

Terminal-Bench Hard

AA · Coding · Objective

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #179 · Source label: GPT-5 nano (Minimal)

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 6.8%
Percentile: 46.4%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `terminalbenchHard`.

46.4% percentile inside its fair comparison set

6.8%Raw benchmark value

SciCode

AA · Coding · Objective

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #101 · Source label: GPT-5.4 nano (Xhigh)

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 47.2%
Percentile: 49.5%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `scicode`.

49.5% percentile inside its fair comparison set

47.2%Raw benchmark value

Text Arena · Coding

AR · Coding · Human

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #121 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,460
Percentile: 67.7%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: coding. Source rank: #136. Votes: 17626. Organization: openai. License: Proprietary.

67.7% percentile inside its fair comparison set

1,460Raw benchmark valueCI 1,454 - 1,466

Text Arena · Coding · No Style Control

AR · Coding · Human

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #150 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,405
Percentile: 59.8%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: coding. Source rank: #166. Votes: 17626. Organization: openai. License: Proprietary.

59.8% percentile inside its fair comparison set

1,405Raw benchmark valueCI 1,399 - 1,410

Vibe Code Bench v1.1

VALS-AI · Coding · Objective

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #59 · Source label: openai/gpt-5.4-nano-2026-03-17

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 26.1%
Percentile: 37.6%
Last updated: recent
Eligibility: headline eligible

Parsed from Vals AI BenchmarkView overall scores. Vals slug: vibe-code; provider: OpenAI.

37.6% percentile inside its fair comparison set

26.1%Raw benchmark valueCI 16.2% - 36%

WebDev Arena

AR · Coding · Human

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #76 · Source label: gpt-5-medium

backfilledproxy backfilledBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,417
Percentile: 41.7%
Last updated: recent
Eligibility: Fallback benchmark identity is visible for context but excluded from default ranking.

Parsed from Arena leaderboard dataset row `gpt-5-medium`. Category: overall. Source rank: #79. Votes: 3091. Organization: openai. License: Proprietary. Backfilled from GPT-5 via approved benchmark identity mapping map-gpt-5-4-nano-to-gpt-5.

41.7% percentile inside its fair comparison set

1,417Raw benchmark valueCI 1,402 - 1,433

Code Arena · Webdev Html

AR · Coding · Human

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #72 · Source label: gpt-5-medium

backfilledproxy backfilledBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,427
Percentile: 44%
Last updated: recent
Eligibility: Fallback benchmark identity is visible for context but excluded from default ranking.

Parsed from Arena leaderboard dataset row `gpt-5-medium`. Category: webdev-html. Source rank: #76. Votes: 3091. Organization: openai. License: Proprietary. Backfilled from GPT-5 via approved benchmark identity mapping map-gpt-5-4-nano-to-gpt-5.

44% percentile inside its fair comparison set

1,427Raw benchmark valueCI 1,410 - 1,443

Code Migration

VALS-AI · Coding · Objective

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #49 · Source label: openai/gpt-5.4-nano-2026-03-17

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 14.5%
Percentile: 29.4%
Last updated: recent
Eligibility: headline eligible

Parsed from Vals AI BenchmarkView overall scores. Vals slug: code-migration; provider: OpenAI.

29.4% percentile inside its fair comparison set

14.5%Raw benchmark valueCI 6.6% - 22.4%

Terminal-Bench 2.1

VALS-AI · Coding · Objective

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #62 · Source label: openai/gpt-5.4-nano-2026-03-17

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 41.6%
Percentile: 9%
Last updated: recent
Eligibility: headline eligible

Parsed from Vals AI BenchmarkView overall scores. Vals slug: terminal-bench-2-1; provider: OpenAI.

9% percentile inside its fair comparison set

41.6%Raw benchmark valueCI 35.7% - 47.4%

SWE-bench Verified

VALS-AI · Coding · Objective

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #60 · Source label: openai/gpt-5.4-nano-2026-03-17

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 69.8%
Percentile: 28%
Last updated: recent
Eligibility: headline eligible

Parsed from Vals AI BenchmarkView overall scores. Vals slug: swebench; provider: OpenAI.

28% percentile inside its fair comparison set

69.8%Raw benchmark valueCI 65.8% - 73.8%

LiveCodeBench

VALS-AI · Coding · Objective

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #40 · Source label: openai/gpt-5.4-nano-2026-03-17

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 84%
Percentile: 68.8%
Last updated: recent
Eligibility: headline eligible

Parsed from Vals AI BenchmarkView overall scores. Vals slug: lcb; provider: OpenAI.

68.8% percentile inside its fair comparison set

84%Raw benchmark valueCI 82% - 86.1%

Agentic coding

LB · Coding · Objective

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #48 · Source label: gpt-5.4-nano-xhigh

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 46.8%
Percentile: 27.7%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Category: Agentic Coding. Tasks scored: 3.

27.7% percentile inside its fair comparison set

46.8%Raw benchmark value

Coding

LB · Coding · Objective

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #58 · Source label: gpt-5.4-nano-xhigh

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 70.8%
Percentile: 12.3%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Category: Coding. Tasks scored: 2.

12.3% percentile inside its fair comparison set

70.8%Raw benchmark value

JavaScript

LB · Coding · Objective

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #32 · Source label: gpt-5.4-nano-xhigh

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 63.6%
Percentile: 58.5%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Task: javascript. Category: Agentic Coding.

58.5% percentile inside its fair comparison set

63.6%Raw benchmark value

TypeScript

LB · Coding · Objective

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #48 · Source label: gpt-5.4-nano-xhigh

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 26.7%
Percentile: 27.7%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Task: typescript. Category: Agentic Coding.

27.7% percentile inside its fair comparison set

26.7%Raw benchmark value

Python

LB · Coding · Objective

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #41 · Source label: gpt-5.4-nano-xhigh

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 50%
Percentile: 46.2%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Task: python. Category: Agentic Coding.

46.2% percentile inside its fair comparison set

50%Raw benchmark value

Coding generation

LB · Coding · Objective

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #51 · Source label: gpt-5.4-nano-xhigh

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 74.3%
Percentile: 23.1%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Task: code_generation. Category: Coding.

23.1% percentile inside its fair comparison set

74.3%Raw benchmark value

Coding completion

LB · Coding · Objective

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #60 · Source label: gpt-5.4-nano-xhigh

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 67.4%
Percentile: 9.2%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Task: code_completion. Category: Coding.

9.2% percentile inside its fair comparison set

67.4%Raw benchmark value

Debugging

BB · Coding · Rubric

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #15 · Source label: openai/gpt-5.4-nano

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: BridgeBench
Raw value: 81.2%
Percentile: 41.7%
Last updated: archived
Eligibility: headline eligible

Parsed from the BridgeBench page for bridgebench-debugging.

41.7% percentile inside its fair comparison set

81.2%Raw benchmark value

Refactoring

BB · Coding · Rubric

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #16 · Source label: openai/gpt-5.4-nano

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: BridgeBench
Raw value: 56.9%
Percentile: 28.6%
Last updated: archived
Eligibility: headline eligible

Parsed from the BridgeBench page for bridgebench-refactoring.

28.6% percentile inside its fair comparison set

56.9%Raw benchmark value

Security

BB · Coding · Rubric

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #12 · Source label: openai/gpt-5.4-nano

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: BridgeBench
Raw value: 80%
Percentile: 56%
Last updated: archived
Eligibility: headline eligible

Parsed from the BridgeBench page for bridgebench-security.

56% percentile inside its fair comparison set

80%Raw benchmark value

UI bench

BB · Coding · Rubric

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #13 · Source label: openai/gpt-5.4-nano

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: BridgeBench
Raw value: 65.3%
Percentile: 45.5%
Last updated: archived
Eligibility: headline eligible

Parsed from the BridgeBench page for bridgebench-ui.

45.5% percentile inside its fair comparison set

65.3%Raw benchmark value

Speed throughput

BB · Coding · Speed / cost

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #5 · Source label: openai/gpt-5.4-nano

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: BridgeBench
Raw value: 227.8 t/s
Percentile: 84%
Last updated: archived
Eligibility: headline eligible

Parsed from the BridgeBench page for bridgebench-throughput.

84% percentile inside its fair comparison set

227.8 t/sRaw benchmark value

Speed TTFT

BB · Coding · Speed / cost

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #7 · Source label: openai/gpt-5.4-nano

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: BridgeBench
Raw value: 941.00ms
Percentile: 76%
Last updated: archived
Eligibility: headline eligible

Parsed from the BridgeBench page for bridgebench-ttft.

76% percentile inside its fair comparison set

941.00msRaw benchmark value

Speed cost

BB · Coding · Speed / cost

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #5 · Source label: openai/gpt-5.4-nano

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: BridgeBench
Raw value: $0 $
Percentile: 84%
Last updated: archived
Eligibility: headline eligible

Parsed from the BridgeBench page for bridgebench-speed-cost.

84% percentile inside its fair comparison set

$0 $Raw benchmark value

Overall quality

BB · Coding · Combined

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #13 · Source label: GPT-5.4 Nano

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: BridgeBench
Raw value: 70.9%
Percentile: 45.5%
Last updated: archived
Eligibility: headline eligible

Parsed from the BridgeBench page for bridgebench-overall-quality.

45.5% percentile inside its fair comparison set

70.9%Raw benchmark value

Overall vibe

BB · Coding · Combined

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #13 · Source label: GPT-5.4 Nano

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: BridgeBench
Raw value: 71.6%
Percentile: 45.5%
Last updated: archived
Eligibility: headline eligible

Parsed from the BridgeBench page for bridgebench-overall-vibe.

45.5% percentile inside its fair comparison set

71.6%Raw benchmark value

Overall coverage

BB · Coding · Combined

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #20 · Source label: GPT-5.4 Nano

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: BridgeBench
Raw value: 85.7%
Percentile: 22.7%
Last updated: archived
Eligibility: headline eligible

Parsed from the BridgeBench page for bridgebench-overall-coverage.

22.7% percentile inside its fair comparison set

85.7%Raw benchmark value

Reasoning / math / science20 benchmarks28.9%

Humanity's Last Exam

AA · Reasoning / math / science · Objective

It is one of the cleaner reads on deliberate reasoning strength rather than style or popularity.

Rank #458 · Source label: GPT-5 nano (Minimal)

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 4%
Percentile: 13.6%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `hle`.

13.6% percentile inside its fair comparison set

4%Raw benchmark value

GPQA

AA · Reasoning / math / science · Objective

It is one of the cleaner reads on deliberate reasoning strength rather than style or popularity.

Rank #386 · Source label: GPT-5 nano (Minimal)

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 42.8%
Percentile: 21.9%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `gpqa`.

21.9% percentile inside its fair comparison set

42.8%Raw benchmark value

CritPt

AA · Reasoning / math / science · Objective

It is one of the cleaner reads on deliberate reasoning strength rather than style or popularity.

Rank #230 · Source label: GPT-5 nano (High)

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 0%
Percentile: 50.1%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `critpt`.

50.1% percentile inside its fair comparison set

0%Raw benchmark value

TutorBench

SL · Reasoning / math / science · Rubric

It is one of the cleaner reads on deliberate reasoning strength rather than style or popularity.

Rank #6 · Source label: gpt-5-2025-08-07

backfilledproxy backfilledBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Scale Labs
Raw value: 55.3%
Percentile: 88.9%
Last updated: recent
Eligibility: Fallback benchmark identity is visible for context but excluded from default ranking.

Parsed from the public Scale Labs page for scale-tutorbench. Reported model configuration: gpt-5-2025-08-07. Collapse policy: highest reported score per canonical model. Backfilled from GPT-5 via approved benchmark identity mapping map-gpt-5-4-nano-to-gpt-5.

88.9% percentile inside its fair comparison set

55.3%Raw benchmark value

MultiNRC

SL · Reasoning / math / science · Rubric

It is one of the cleaner reads on deliberate reasoning strength rather than style or popularity.

Rank #4 · Source label: gpt-5-pro-2025-10-06

backfilledproxy backfilledBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Scale Labs
Raw value: 65.2%
Percentile: 97.7%
Last updated: recent
Eligibility: Fallback benchmark identity is visible for context but excluded from default ranking.

Parsed from the public Scale Labs page for scale-multinrc. Reported model configuration: gpt-5-pro-2025-10-06. Collapse policy: highest reported score per canonical model. Backfilled from GPT-5 via approved benchmark identity mapping map-gpt-5-4-nano-to-gpt-5.

97.7% percentile inside its fair comparison set

65.2%Raw benchmark value

Text Arena · Math

AR · Reasoning / math / science · Human

It is one of the cleaner reads on deliberate reasoning strength rather than style or popularity.

Rank #106 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,425
Percentile: 70.8%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: math. Source rank: #117. Votes: 3257. Organization: openai. License: Proprietary.

70.8% percentile inside its fair comparison set

1,425Raw benchmark valueCI 1,414 - 1,436

Text Arena · Math · No Style Control

AR · Reasoning / math / science · Human

It is one of the cleaner reads on deliberate reasoning strength rather than style or popularity.

Rank #133 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,406
Percentile: 63.2%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: math. Source rank: #146. Votes: 3257. Organization: openai. License: Proprietary.

63.2% percentile inside its fair comparison set

1,406Raw benchmark valueCI 1,395 - 1,417

GPQA Diamond

VALS-AI · Reasoning / math / science · Objective

It is one of the cleaner reads on deliberate reasoning strength rather than style or popularity.

Rank #76 · Source label: openai/gpt-5.4-nano-2026-03-17

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 77.5%
Percentile: 39.5%
Last updated: recent
Eligibility: headline eligible

Parsed from Vals AI BenchmarkView overall scores. Vals slug: gpqa; provider: OpenAI.

39.5% percentile inside its fair comparison set

77.5%Raw benchmark valueCI 72.8% - 82.3%

MMLU Pro

VALS-AI · Reasoning / math / science · Objective

It is one of the cleaner reads on deliberate reasoning strength rather than style or popularity.

Rank #102 · Source label: openai/gpt-5.4-nano-2026-03-17

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 77.2%
Percentile: 19.2%
Last updated: recent
Eligibility: headline eligible

Parsed from Vals AI BenchmarkView overall scores. Vals slug: mmlu_pro; provider: OpenAI.

19.2% percentile inside its fair comparison set

77.2%Raw benchmark valueCI 76.3% - 78%

Mathematics

LB · Reasoning / math / science · Objective

It is one of the cleaner reads on deliberate reasoning strength rather than style or popularity.

Rank #30 · Source label: gpt-5.4-nano-xhigh

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 91%
Percentile: 55.4%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Category: Mathematics. Tasks scored: 4.

55.4% percentile inside its fair comparison set

91%Raw benchmark value

Reasoning

LB · Reasoning / math / science · Objective

It is one of the cleaner reads on deliberate reasoning strength rather than style or popularity.

Rank #49 · Source label: gpt-5.4-nano-xhigh

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 81.1%
Percentile: 26.2%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Category: Reasoning. Tasks scored: 4.

26.2% percentile inside its fair comparison set

81.1%Raw benchmark value

AMPS Hard

LB · Reasoning / math / science · Objective

It is one of the cleaner reads on deliberate reasoning strength rather than style or popularity.

Rank #20 · Source label: gpt-5.4-nano-xhigh

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 98%
Percentile: 83.1%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Task: AMPS_Hard. Category: Mathematics.

83.1% percentile inside its fair comparison set

98%Raw benchmark value

Integrals with game

LB · Reasoning / math / science · Objective

It is one of the cleaner reads on deliberate reasoning strength rather than style or popularity.

Rank #29 · Source label: gpt-5.4-nano-xhigh

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 82.8%
Percentile: 56.9%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Task: integrals_with_game. Category: Mathematics.

56.9% percentile inside its fair comparison set

82.8%Raw benchmark value

Math competition

LB · Reasoning / math / science · Objective

It is one of the cleaner reads on deliberate reasoning strength rather than style or popularity.

Rank #40 · Source label: gpt-5.4-nano-xhigh

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 95.1%
Percentile: 49.2%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Task: math_comp. Category: Mathematics.

49.2% percentile inside its fair comparison set

95.1%Raw benchmark value

Olympiad

LB · Reasoning / math / science · Objective

It is one of the cleaner reads on deliberate reasoning strength rather than style or popularity.

Rank #48 · Source label: gpt-5.4-nano-xhigh

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 88%
Percentile: 27.7%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Task: olympiad. Category: Mathematics.

27.7% percentile inside its fair comparison set

88%Raw benchmark value

Theory of mind

LB · Reasoning / math / science · Objective

It is one of the cleaner reads on deliberate reasoning strength rather than style or popularity.

Rank #28 · Source label: gpt-5.4-nano-xhigh

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 80.8%
Percentile: 66.2%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Task: theory_of_mind. Category: Reasoning.

66.2% percentile inside its fair comparison set

80.8%Raw benchmark value

Zebra puzzle

LB · Reasoning / math / science · Objective

It is one of the cleaner reads on deliberate reasoning strength rather than style or popularity.

Rank #28 · Source label: gpt-5.4-nano-xhigh

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 97.6%
Percentile: 58.5%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Task: zebra_puzzle. Category: Reasoning.

58.5% percentile inside its fair comparison set

97.6%Raw benchmark value

Spatial

LB · Reasoning / math / science · Objective

It is one of the cleaner reads on deliberate reasoning strength rather than style or popularity.

Rank #66 · Source label: gpt-5.4-nano-xhigh

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 82%
Percentile: 0%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Task: spatial. Category: Reasoning.

0% percentile inside its fair comparison set

82%Raw benchmark value

Logic with navigation

LB · Reasoning / math / science · Objective

It is one of the cleaner reads on deliberate reasoning strength rather than style or popularity.

Rank #61 · Source label: gpt-5.4-nano-xhigh

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 64%
Percentile: 10.8%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Task: logic_with_navigation. Category: Reasoning.

10.8% percentile inside its fair comparison set

64%Raw benchmark value

Reasoning

BB · Reasoning / math / science · Rubric

It is one of the cleaner reads on deliberate reasoning strength rather than style or popularity.

Rank #19 · Source label: openai/gpt-5.4-nano

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: BridgeBench
Raw value: 34.8%
Percentile: 14.3%
Last updated: archived
Eligibility: headline eligible

Parsed from the BridgeBench page for bridgebench-reasoning.

14.3% percentile inside its fair comparison set

34.8%Raw benchmark value

Professional reasoning35 benchmarks36%

APEX-Agents-AA

AA · Professional reasoning · Objective

Long-horizon agentic task completion.

Rank #18 · Source label: GPT-5.4 nano (Xhigh)

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 24.9%
Percentile: 46.9%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `apexAgents`.

46.9% percentile inside its fair comparison set

24.9%Raw benchmark value

PRBench Legal

SL · Professional reasoning · Rubric

Applied legal reasoning on professional-domain tasks.

Rank #10 · Source label: gpt-5-pro

backfilledproxy backfilledBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Scale Labs
Raw value: 49.9%
Percentile: 82.9%
Last updated: recent
Eligibility: Fallback benchmark identity is visible for context but excluded from default ranking.

Parsed from the public Scale Labs page for scale-prbench-legal. Reported model configuration: gpt-5-pro. Collapse policy: highest reported score per canonical model. Backfilled from GPT-5 via approved benchmark identity mapping map-gpt-5-4-nano-to-gpt-5.

82.9% percentile inside its fair comparison set

49.9%Raw benchmark value

Text Arena · Expert

AR · Professional reasoning · Human

Observed user preference in Arena's Text Arena expert leaderboard.

Rank #121 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,437
Percentile: 63.3%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: expert. Source rank: #134. Votes: 6344. Organization: openai. License: Proprietary.

63.3% percentile inside its fair comparison set

1,437Raw benchmark valueCI 1,429 - 1,446

Text Arena · Industry Business And Management And Financial Operations

AR · Professional reasoning · Human

Observed user preference in Arena's Text Arena industry_business_and_management_and_financial_operations leaderboard.

Rank #137 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,405
Percentile: 63.1%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: industry_business_and_management_and_financial_operations. Source rank: #151. Votes: 12559. Organization: openai. License: Proprietary.

63.1% percentile inside its fair comparison set

1,405Raw benchmark valueCI 1,399 - 1,412

Text Arena · Industry Entertainment And Sports And Media

AR · Professional reasoning · Human

Observed user preference in Arena's Text Arena industry_entertainment_and_sports_and_media leaderboard.

Rank #153 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,352
Percentile: 59.4%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: industry_entertainment_and_sports_and_media. Source rank: #172. Votes: 13657. Organization: openai. License: Proprietary.

59.4% percentile inside its fair comparison set

1,352Raw benchmark valueCI 1,345 - 1,358

Text Arena · Industry Legal And Government

AR · Professional reasoning · Human

Observed user preference in Arena's Text Arena industry_legal_and_government leaderboard.

Rank #150 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,400
Percentile: 57.3%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: industry_legal_and_government. Source rank: #168. Votes: 4954. Organization: openai. License: Proprietary.

57.3% percentile inside its fair comparison set

1,400Raw benchmark valueCI 1,391 - 1,409

Text Arena · Industry Life And Physical And Social Science

AR · Professional reasoning · Human

Observed user preference in Arena's Text Arena industry_life_and_physical_and_social_science leaderboard.

Rank #145 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,415
Percentile: 61.5%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: industry_life_and_physical_and_social_science. Source rank: #162. Votes: 10492. Organization: openai. License: Proprietary.

61.5% percentile inside its fair comparison set

1,415Raw benchmark valueCI 1,408 - 1,422

Text Arena · Industry Mathematical

AR · Professional reasoning · Human

Observed user preference in Arena's Text Arena industry_mathematical leaderboard.

Rank #109 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,431
Percentile: 69.7%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: industry_mathematical. Source rank: #119. Votes: 3478. Organization: openai. License: Proprietary.

69.7% percentile inside its fair comparison set

1,431Raw benchmark valueCI 1,420 - 1,442

Text Arena · Industry Medicine And Healthcare

AR · Professional reasoning · Human

Observed user preference in Arena's Text Arena industry_medicine_and_healthcare leaderboard.

Rank #150 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,413
Percentile: 56.9%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: industry_medicine_and_healthcare. Source rank: #168. Votes: 4788. Organization: openai. License: Proprietary.

56.9% percentile inside its fair comparison set

1,413Raw benchmark valueCI 1,403 - 1,423

Text Arena · Industry Software And It Services

AR · Professional reasoning · Human

Observed user preference in Arena's Text Arena industry_software_and_it_services leaderboard.

Rank #125 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,450
Percentile: 67%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: industry_software_and_it_services. Source rank: #139. Votes: 24913. Organization: openai. License: Proprietary.

67% percentile inside its fair comparison set

1,450Raw benchmark valueCI 1,445 - 1,455

Text Arena · Industry Writing And Literature And Language

AR · Professional reasoning · Human

Observed user preference in Arena's Text Arena industry_writing_and_literature_and_language leaderboard.

Rank #159 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,363
Percentile: 57.9%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: industry_writing_and_literature_and_language. Source rank: #179. Votes: 15381. Organization: openai. License: Proprietary.

57.9% percentile inside its fair comparison set

1,363Raw benchmark valueCI 1,357 - 1,369

Text Arena · Expert · No Style Control

AR · Professional reasoning · Human

Observed user preference in Arena's Text Arena expert leaderboard.

Rank #142 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,396
Percentile: 56.9%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: expert. Source rank: #159. Votes: 6344. Organization: openai. License: Proprietary.

56.9% percentile inside its fair comparison set

1,396Raw benchmark valueCI 1,388 - 1,405

Text Arena · Industry Business And Management And Financial Operations · No Style Control

AR · Professional reasoning · Human

Observed user preference in Arena's Text Arena industry_business_and_management_and_financial_operations leaderboard.

Rank #158 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,366
Percentile: 57.5%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: industry_business_and_management_and_financial_operations. Source rank: #175. Votes: 12559. Organization: openai. License: Proprietary.

57.5% percentile inside its fair comparison set

1,366Raw benchmark valueCI 1,360 - 1,373

Text Arena · Industry Entertainment And Sports And Media · No Style Control

AR · Professional reasoning · Human

Observed user preference in Arena's Text Arena industry_entertainment_and_sports_and_media leaderboard.

Rank #172 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,324
Percentile: 54.3%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: industry_entertainment_and_sports_and_media. Source rank: #193. Votes: 13657. Organization: openai. License: Proprietary.

54.3% percentile inside its fair comparison set

1,324Raw benchmark valueCI 1,317 - 1,331

Text Arena · Industry Legal And Government · No Style Control

AR · Professional reasoning · Human

Observed user preference in Arena's Text Arena industry_legal_and_government leaderboard.

Rank #167 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,367
Percentile: 52.4%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: industry_legal_and_government. Source rank: #187. Votes: 4954. Organization: openai. License: Proprietary.

52.4% percentile inside its fair comparison set

1,367Raw benchmark valueCI 1,358 - 1,377

Text Arena · Industry Life And Physical And Social Science · No Style Control

AR · Professional reasoning · Human

Observed user preference in Arena's Text Arena industry_life_and_physical_and_social_science leaderboard.

Rank #170 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,374
Percentile: 54.8%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: industry_life_and_physical_and_social_science. Source rank: #187. Votes: 10492. Organization: openai. License: Proprietary.

54.8% percentile inside its fair comparison set

1,374Raw benchmark valueCI 1,367 - 1,381

Text Arena · Industry Mathematical · No Style Control

AR · Professional reasoning · Human

Observed user preference in Arena's Text Arena industry_mathematical leaderboard.

Rank #138 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,408
Percentile: 61.5%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: industry_mathematical. Source rank: #150. Votes: 3478. Organization: openai. License: Proprietary.

61.5% percentile inside its fair comparison set

1,408Raw benchmark valueCI 1,398 - 1,419

Text Arena · Industry Medicine And Healthcare · No Style Control

AR · Professional reasoning · Human

Observed user preference in Arena's Text Arena industry_medicine_and_healthcare leaderboard.

Rank #164 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,369
Percentile: 52.9%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: industry_medicine_and_healthcare. Source rank: #183. Votes: 4788. Organization: openai. License: Proprietary.

52.9% percentile inside its fair comparison set

1,369Raw benchmark valueCI 1,359 - 1,378

Text Arena · Industry Software And It Services · No Style Control

AR · Professional reasoning · Human

Observed user preference in Arena's Text Arena industry_software_and_it_services leaderboard.

Rank #154 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,402
Percentile: 59.3%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: industry_software_and_it_services. Source rank: #171. Votes: 24913. Organization: openai. License: Proprietary.

59.3% percentile inside its fair comparison set

1,402Raw benchmark valueCI 1,397 - 1,408

Text Arena · Industry Writing And Literature And Language · No Style Control

AR · Professional reasoning · Human

Observed user preference in Arena's Text Arena industry_writing_and_literature_and_language leaderboard.

Rank #170 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,338
Percentile: 54.9%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: industry_writing_and_literature_and_language. Source rank: #192. Votes: 15381. Organization: openai. License: Proprietary.

54.9% percentile inside its fair comparison set

1,338Raw benchmark valueCI 1,332 - 1,344

Legal Research Bench

VALS-AI · Professional reasoning · Objective

Applied legal research tasks.

Rank #64 · Source label: openai/gpt-5.4-nano-2026-03-17

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 6.3%
Percentile: 7.4%
Last updated: recent
Eligibility: headline eligible

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_research; provider: OpenAI.

7.4% percentile inside its fair comparison set

6.3%Raw benchmark valueCI 3% - 9.5%

Harvey's Legal Agent Benchmark

VALS-AI · Professional reasoning · Objective

Completing legal work with documents, spreadsheets, presentations, and file-system tools.

Rank #60 · Source label: openai/gpt-5.4-nano-2026-03-17

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 0%
Percentile: 20.3%
Last updated: recent
Eligibility: headline eligible

Parsed from Vals AI BenchmarkView overall scores. Vals slug: hlab; provider: OpenAI.

20.3% percentile inside its fair comparison set

0%Raw benchmark valueCI 0% - 0%

LegalBench

VALS-AI · Professional reasoning · Objective

Academic legal reasoning tasks.

Rank #97 · Source label: openai/gpt-5.4-nano-2026-03-17

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 77.9%
Percentile: 28.9%
Last updated: recent
Eligibility: headline eligible

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: OpenAI.

28.9% percentile inside its fair comparison set

77.9%Raw benchmark valueCI 77.1% - 78.8%

Finance Agent v2

VALS-AI · Professional reasoning · Objective

Core financial analyst tasks for agentic models.

Rank #56 · Source label: openai/gpt-5.4-nano-2026-03-17

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 38.2%
Percentile: 20.3%
Last updated: recent
Eligibility: headline eligible

Parsed from Vals AI BenchmarkView overall scores. Vals slug: fabv2; provider: OpenAI.

20.3% percentile inside its fair comparison set

38.2%Raw benchmark valueCI 35.9% - 40.5%

MedCode

VALS-AI · Professional reasoning · Objective

Medical billing support and coding tasks.

Rank #58 · Source label: openai/gpt-5.4-nano-2026-03-17

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 41%
Percentile: 40%
Last updated: recent
Eligibility: headline eligible

Parsed from Vals AI BenchmarkView overall scores. Vals slug: medcode; provider: OpenAI.

40% percentile inside its fair comparison set

41%Raw benchmark valueCI 36.6% - 45.5%

MedScribe

VALS-AI · Professional reasoning · Objective

Administrative documentation support for doctors.

Rank #64 · Source label: openai/gpt-5.4-nano-2026-03-17

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 77.1%
Percentile: 34.4%
Last updated: recent
Eligibility: headline eligible

Parsed from Vals AI BenchmarkView overall scores. Vals slug: medscribe; provider: OpenAI.

34.4% percentile inside its fair comparison set

77.1%Raw benchmark valueCI 73.4% - 80.8%

SAGE

VALS-AI · Professional reasoning · Objective

Student Assessment with Generative Evaluation.

Rank #60 · Source label: openai/gpt-5.4-nano-2026-03-17

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 38.1%
Percentile: 27.2%
Last updated: recent
Eligibility: headline eligible

Parsed from Vals AI BenchmarkView overall scores. Vals slug: sage; provider: OpenAI.

27.2% percentile inside its fair comparison set

38.1%Raw benchmark valueCI 32% - 44.1%

TaxEval v2

VALS-AI · Professional reasoning · Objective

Answer quality on tax questions and responses.

Rank #100 · Source label: openai/gpt-5.4-nano-2026-03-17

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 67.4%
Percentile: 23.3%
Last updated: recent
Eligibility: headline eligible

Parsed from Vals AI BenchmarkView overall scores. Vals slug: tax_eval_v2; provider: OpenAI.

23.3% percentile inside its fair comparison set

67.4%Raw benchmark valueCI 65.6% - 69.2%

Data analysis

LB · Professional reasoning · Objective

Structured data manipulation and table reasoning accuracy.

Rank #56 · Source label: gpt-5.4-nano-xhigh

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 67.6%
Percentile: 15.4%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Category: Data Analysis. Tasks scored: 3.

15.4% percentile inside its fair comparison set

67.6%Raw benchmark value

Overall

LB · Professional reasoning · Objective

Average objective performance across LiveBench's current public category mix.

Rank #55 · Source label: gpt-5.4-nano-xhigh

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 69.6%
Percentile: 16.9%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Category averages included: 7.

16.9% percentile inside its fair comparison set

69.6%Raw benchmark value

Consecutive events

LB · Professional reasoning · Objective

Objective consecutive events score in LiveBench.

Rank #56 · Source label: gpt-5.4-nano-xhigh

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 54.3%
Percentile: 15.4%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Task: consecutive_events. Category: Data Analysis.

15.4% percentile inside its fair comparison set

54.3%Raw benchmark value

Table join

LB · Professional reasoning · Objective

Objective table join score in LiveBench.

Rank #18 · Source label: gpt-5.4-nano-xhigh

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 50.6%
Percentile: 73.8%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Task: tablejoin. Category: Data Analysis.

73.8% percentile inside its fair comparison set

50.6%Raw benchmark value

Table reformat

LB · Professional reasoning · Objective

Objective table reformat score in LiveBench.

Rank #32 · Source label: gpt-5.4-nano-xhigh

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 98%
Percentile: 61.5%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Task: tablereformat. Category: Data Analysis.

61.5% percentile inside its fair comparison set

98%Raw benchmark value

Hallucination

BB · Professional reasoning · Rubric

Accuracy and fabrication resistance under prompts that invite unsupported claims.

Rank #23 · Source label: openai/gpt-5.4-nano

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: BridgeBench
Raw value: 69.6%
Percentile: 35.3%
Last updated: archived
Eligibility: headline eligible

Parsed from the BridgeBench page for bridgebench-hallucination.

35.3% percentile inside its fair comparison set

69.6%Raw benchmark value

Poker Agent

VALS-AI · Professional reasoning · Objective

Agent profit in poker-style strategic play.

Rank #5 · Source label: openai/gpt-5-2025-08-07

backfilledproxy backfilledBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 1,103.2 score
Percentile: 94.7%
Last updated: archived
Eligibility: Fallback benchmark identity is visible for context but excluded from default ranking.

Parsed from Vals AI BenchmarkView overall scores. Vals slug: poker_agent; provider: unknown. Backfilled from GPT-5 via approved benchmark identity mapping map-gpt-5-4-nano-to-gpt-5.

94.7% percentile inside its fair comparison set

1,103.2 scoreRaw benchmark valueCI 1,103.2 score - 1,103.2 score

Search / tool use3 benchmarks37.2%

Tau2-Bench Telecom

AA · Search / tool use · Objective

It matters when the model must browse, call tools, and recover useful answers from external systems.

Rank #214 · Source label: GPT-5 nano (Minimal)

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 25.7%
Percentile: 37.2%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `tau2`.

37.2% percentile inside its fair comparison set

25.7%Raw benchmark value

Search Arena

AR · Search / tool use · Human

It matters when the model must browse, call tools, and recover useful answers from external systems.

Rank #23 · Source label: gpt-5-search

backfilledproxy backfilledBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,183
Percentile: 39.4%
Last updated: recent
Eligibility: Fallback benchmark identity is visible for context but excluded from default ranking.

Parsed from Arena leaderboard dataset row `gpt-5-search`. Category: overall. Source rank: #22. Votes: 20781. Organization: openai. License: Proprietary. Backfilled from GPT-5 via approved benchmark identity mapping map-gpt-5-4-nano-to-gpt-5.

39.4% percentile inside its fair comparison set

1,183Raw benchmark valueCI 1,175 - 1,190

Search Arena · No Style Control

AR · Search / tool use · Human

It matters when the model must browse, call tools, and recover useful answers from external systems.

Rank #30 · Source label: gpt-5-search

backfilledproxy backfilledBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,133
Percentile: 18.2%
Last updated: recent
Eligibility: Fallback benchmark identity is visible for context but excluded from default ranking.

Parsed from Arena leaderboard dataset row `gpt-5-search`. Category: overall. Source rank: #30. Votes: 20781. Organization: openai. License: Proprietary. Backfilled from GPT-5 via approved benchmark identity mapping map-gpt-5-4-nano-to-gpt-5.

18.2% percentile inside its fair comparison set

1,133Raw benchmark valueCI 1,127 - 1,139

Long context2 benchmarks38.2%

Long Context Reasoning

AA · Long context · Objective

It checks whether long-context claims survive contact with retrieval, memory, or long-document tasks.

Rank #342 · Source label: GPT-5 nano (Minimal)

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 20%
Percentile: 25.1%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `lcr`.

25.1% percentile inside its fair comparison set

20%Raw benchmark value

CorpFin v2

VALS-AI · Long context · Objective

It checks whether long-context claims survive contact with retrieval, memory, or long-document tasks.

Rank #58 · Source label: openai/gpt-5.4-nano-2026-03-17

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 61.2%
Percentile: 51.3%
Last updated: aging
Eligibility: headline eligible

Parsed from Vals AI BenchmarkView overall scores. Vals slug: corp_fin_v2; provider: OpenAI.

51.3% percentile inside its fair comparison set

61.2%Raw benchmark valueCI 59.3% - 63.1%

Vision understanding24 benchmarks26.2%

MMMU-Pro

AA · Vision understanding · Objective

It is useful when the model must read charts, UI, screenshots, or visual scenes rather than text alone.

Rank #201 · Source label: GPT-5 nano (Minimal)

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 31.8%
Percentile: 3.8%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

3.8% percentile inside its fair comparison set

31.8%Raw benchmark value

VISTA

SL · Vision understanding · Rubric

It is useful when the model must read charts, UI, screenshots, or visual scenes rather than text alone.

Rank #6 · Source label: gpt-5-pro-2025-10-06

backfilledproxy backfilledBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Scale Labs
Raw value: 52.4%
Percentile: 95.2%
Last updated: recent
Eligibility: Fallback benchmark identity is visible for context but excluded from default ranking.

Parsed from the public Scale Labs page for scale-vista. Reported model configuration: gpt-5-pro-2025-10-06. Collapse policy: highest reported score per canonical model. Backfilled from GPT-5 via approved benchmark identity mapping map-gpt-5-4-nano-to-gpt-5.

95.2% percentile inside its fair comparison set

52.4%Raw benchmark value

VTB

SL · Vision understanding · Rubric

It is useful when the model must read charts, UI, screenshots, or visual scenes rather than text alone.

Rank #8 · Source label: gpt-5-2025-08-07-thinking

backfilledproxy backfilledBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Scale Labs
Raw value: 18.7%
Percentile: 79.2%
Last updated: recent
Eligibility: Fallback benchmark identity is visible for context but excluded from default ranking.

Parsed from the public Scale Labs page for scale-vtb. Reported model configuration: gpt-5-2025-08-07-thinking. Collapse policy: highest reported score per canonical model. Backfilled from GPT-5 via approved benchmark identity mapping map-gpt-5-4-nano-to-gpt-5.

79.2% percentile inside its fair comparison set

18.7%Raw benchmark value

Vision Arena

AR · Vision understanding · Human

It is useful when the model must read charts, UI, screenshots, or visual scenes rather than text alone.

Rank #78 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,199
Percentile: 44.6%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: overall. Source rank: #87. Votes: 25647. Organization: openai. License: Proprietary.

44.6% percentile inside its fair comparison set

1,199Raw benchmark valueCI 1,193 - 1,206

Vision Arena · Captioning

AR · Vision understanding · Human

It is useful when the model must read charts, UI, screenshots, or visual scenes rather than text alone.

Rank #16 · Source label: gpt-5-chat

backfilledproxy backfilledBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,195
Percentile: 58.1%
Last updated: recent
Eligibility: Fallback benchmark identity is visible for context but excluded from default ranking.

Parsed from Arena leaderboard dataset row `gpt-5-chat`. Category: captioning. Source rank: #15. Votes: 375. Organization: openai. License: Proprietary. Backfilled from GPT-5 via approved benchmark identity mapping map-gpt-5-4-nano-to-gpt-5.

58.1% percentile inside its fair comparison set

1,195Raw benchmark valueCI 1,163 - 1,227

Vision Arena · Creative Writing Vision

AR · Vision understanding · Human

It is useful when the model must read charts, UI, screenshots, or visual scenes rather than text alone.

Rank #77 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,156
Percentile: 7.3%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: creative_writing_vision. Source rank: #88. Votes: 1530. Organization: openai. License: Proprietary.

7.3% percentile inside its fair comparison set

1,156Raw benchmark valueCI 1,140 - 1,173

Vision Arena · Diagram

AR · Vision understanding · Human

It is useful when the model must read charts, UI, screenshots, or visual scenes rather than text alone.

Rank #73 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,227
Percentile: 27.3%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: diagram. Source rank: #82. Votes: 6845. Organization: openai. License: Proprietary.

27.3% percentile inside its fair comparison set

1,227Raw benchmark valueCI 1,218 - 1,236

Vision Arena · English

AR · Vision understanding · Human

It is useful when the model must read charts, UI, screenshots, or visual scenes rather than text alone.

Rank #74 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,204
Percentile: 47.1%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: english. Source rank: #84. Votes: 10343. Organization: openai. License: Proprietary.

47.1% percentile inside its fair comparison set

1,204Raw benchmark valueCI 1,195 - 1,212

Vision Arena · Entity Recognition

AR · Vision understanding · Human

It is useful when the model must read charts, UI, screenshots, or visual scenes rather than text alone.

Rank #41 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,119
Percentile: 0%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: entity_recognition. Source rank: #50. Votes: 175. Organization: openai. License: Proprietary.

0% percentile inside its fair comparison set

1,119Raw benchmark valueCI 1,077 - 1,162

Vision Arena · Homework

AR · Vision understanding · Human

It is useful when the model must read charts, UI, screenshots, or visual scenes rather than text alone.

Rank #70 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,233
Percentile: 28.1%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: homework. Source rank: #81. Votes: 3447. Organization: openai. License: Proprietary.

28.1% percentile inside its fair comparison set

1,233Raw benchmark valueCI 1,222 - 1,244

Vision Arena · Humor

AR · Vision understanding · Human

It is useful when the model must read charts, UI, screenshots, or visual scenes rather than text alone.

Rank #67 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,155
Percentile: 9.6%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: humor. Source rank: #78. Votes: 922. Organization: openai. License: Proprietary.

9.6% percentile inside its fair comparison set

1,155Raw benchmark valueCI 1,134 - 1,176

Vision Arena · Ocr

AR · Vision understanding · Human

It is useful when the model must read charts, UI, screenshots, or visual scenes rather than text alone.

Rank #76 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,213
Percentile: 25%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: ocr. Source rank: #85. Votes: 18243. Organization: openai. License: Proprietary.

25% percentile inside its fair comparison set

1,213Raw benchmark valueCI 1,207 - 1,220

Vision Arena · No Style Control

AR · Vision understanding · Human

It is useful when the model must read charts, UI, screenshots, or visual scenes rather than text alone.

Rank #77 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,196
Percentile: 45.3%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: overall. Source rank: #88. Votes: 25647. Organization: openai. License: Proprietary.

45.3% percentile inside its fair comparison set

1,196Raw benchmark valueCI 1,190 - 1,202

Vision Arena · Captioning · No Style Control

AR · Vision understanding · Human

It is useful when the model must read charts, UI, screenshots, or visual scenes rather than text alone.

Rank #20 · Source label: gpt-5-chat

backfilledproxy backfilledBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,212
Percentile: 45.2%
Last updated: recent
Eligibility: Fallback benchmark identity is visible for context but excluded from default ranking.

Parsed from Arena leaderboard dataset row `gpt-5-chat`. Category: captioning. Source rank: #19. Votes: 375. Organization: openai. License: Proprietary. Backfilled from GPT-5 via approved benchmark identity mapping map-gpt-5-4-nano-to-gpt-5.

45.2% percentile inside its fair comparison set

1,212Raw benchmark valueCI 1,181 - 1,244

Vision Arena · Creative Writing Vision · No Style Control

AR · Vision understanding · Human

It is useful when the model must read charts, UI, screenshots, or visual scenes rather than text alone.

Rank #77 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,157
Percentile: 7.3%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: creative_writing_vision. Source rank: #89. Votes: 1530. Organization: openai. License: Proprietary.

7.3% percentile inside its fair comparison set

1,157Raw benchmark valueCI 1,141 - 1,174

Vision Arena · Diagram · No Style Control

AR · Vision understanding · Human

It is useful when the model must read charts, UI, screenshots, or visual scenes rather than text alone.

Rank #73 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,211
Percentile: 27.3%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: diagram. Source rank: #83. Votes: 6845. Organization: openai. License: Proprietary.

27.3% percentile inside its fair comparison set

1,211Raw benchmark valueCI 1,202 - 1,220

Vision Arena · English · No Style Control

AR · Vision understanding · Human

It is useful when the model must read charts, UI, screenshots, or visual scenes rather than text alone.

Rank #76 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,201
Percentile: 45.7%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: english. Source rank: #87. Votes: 10343. Organization: openai. License: Proprietary.

45.7% percentile inside its fair comparison set

1,201Raw benchmark valueCI 1,193 - 1,210

Vision Arena · Entity Recognition · No Style Control

AR · Vision understanding · Human

It is useful when the model must read charts, UI, screenshots, or visual scenes rather than text alone.

Rank #40 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,143
Percentile: 2.5%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: entity_recognition. Source rank: #49. Votes: 175. Organization: openai. License: Proprietary.

2.5% percentile inside its fair comparison set

1,143Raw benchmark valueCI 1,101 - 1,186

Vision Arena · Homework · No Style Control

AR · Vision understanding · Human

It is useful when the model must read charts, UI, screenshots, or visual scenes rather than text alone.

Rank #73 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,225
Percentile: 25%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: homework. Source rank: #83. Votes: 3447. Organization: openai. License: Proprietary.

25% percentile inside its fair comparison set

1,225Raw benchmark valueCI 1,214 - 1,236

Vision Arena · Humor · No Style Control

AR · Vision understanding · Human

It is useful when the model must read charts, UI, screenshots, or visual scenes rather than text alone.

Rank #69 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,160
Percentile: 6.8%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: humor. Source rank: #81. Votes: 922. Organization: openai. License: Proprietary.

6.8% percentile inside its fair comparison set

1,160Raw benchmark valueCI 1,139 - 1,181

Vision Arena · Ocr · No Style Control

AR · Vision understanding · Human

It is useful when the model must read charts, UI, screenshots, or visual scenes rather than text alone.

Rank #74 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,204
Percentile: 27%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: ocr. Source rank: #84. Votes: 18243. Organization: openai. License: Proprietary.

27% percentile inside its fair comparison set

1,204Raw benchmark valueCI 1,198 - 1,211

MMMU Pro

VALS-AI · Vision understanding · Objective

It is useful when the model must read charts, UI, screenshots, or visual scenes rather than text alone.

Rank #59 · Source label: openai/gpt-5.4-nano-2026-03-17

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 73.6%
Percentile: 30.1%
Last updated: recent
Eligibility: headline eligible

Parsed from Vals AI BenchmarkView overall scores. Vals slug: mmmu; provider: OpenAI.

30.1% percentile inside its fair comparison set

73.6%Raw benchmark valueCI 71.5% - 75.7%

Vision Arena · Creative Writing

AR · Vision understanding · Human

It is useful when the model must read charts, UI, screenshots, or visual scenes rather than text alone.

Rank #33 · Source label: gpt-5-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,092
Percentile: 0%
Last updated: archived
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5-nano-high`. Category: creative_writing. Source rank: #34. Votes: 206. Organization: openai. License: Proprietary.

0% percentile inside its fair comparison set

1,092Raw benchmark valueCI 1,050 - 1,134

Vision Arena · Creative Writing · No Style Control

AR · Vision understanding · Human

It is useful when the model must read charts, UI, screenshots, or visual scenes rather than text alone.

Rank #32 · Source label: gpt-5-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,117
Percentile: 3.1%
Last updated: archived
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5-nano-high`. Category: creative_writing. Source rank: #33. Votes: 206. Organization: openai. License: Proprietary.

3.1% percentile inside its fair comparison set

1,117Raw benchmark valueCI 1,077 - 1,158

Document understanding3 benchmarks20%

MortgageTax

VALS-AI · Document understanding · Objective

It matters when the job is reading PDFs, tables, forms, or mixed-layout documents rather than plain chat.

Rank #66 · Source label: openai/gpt-5.4-nano-2026-03-17

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 59.1%
Percentile: 24.4%
Last updated: recent
Eligibility: headline eligible

Parsed from Vals AI BenchmarkView overall scores. Vals slug: mortgage_tax; provider: OpenAI.

24.4% percentile inside its fair comparison set

59.1%Raw benchmark valueCI 57.2% - 61%

Vals Multimodal Index

VALS-AI · Document understanding · Combined

It matters when the job is reading PDFs, tables, forms, or mixed-layout documents rather than plain chat.

Rank #28 · Source label: openai/gpt-5.4-nano-2026-03-17

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 48
Percentile: 15.6%
Last updated: aging
Eligibility: headline eligible

Parsed from Vals AI BenchmarkView overall scores. Vals slug: vals_multimodal_index; provider: OpenAI.

15.6% percentile inside its fair comparison set

48Raw benchmark valueCI 44 - 51

Multimodal mix

OC · Document understanding · Objective

It matters when the job is reading PDFs, tables, forms, or mixed-layout documents rather than plain chat.

Rank #5 · Source label: gpt-5

backfilledproxy backfilledBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: OpenCompass
Raw value: 75.4%
Percentile: 92.9%
Last updated: archived
Eligibility: Fallback benchmark identity is visible for context but excluded from default ranking.

Backfilled from GPT-5 via approved benchmark identity mapping map-gpt-5-4-nano-to-gpt-5.

92.9% percentile inside its fair comparison set

75.4%Raw benchmark value

Safety1 benchmarkn/a

MASK

SL · Safety · Rubric

Whether a model stays honest instead of covertly optimizing against the user.

Rank #20 · Source label: gpt-5-pro-2025-10-06

backfilledproxy backfilledBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Scale Labs
Raw value: 86%
Percentile: 74.6%
Last updated: recent
Eligibility: Fallback benchmark identity is visible for context but excluded from default ranking.

Parsed from the public Scale Labs page for scale-mask. Reported model configuration: gpt-5-pro-2025-10-06. Collapse policy: highest reported score per canonical model. Backfilled from GPT-5 via approved benchmark identity mapping map-gpt-5-4-nano-to-gpt-5.

74.6% percentile inside its fair comparison set

86%Raw benchmark value

Embeddings / retrieval1 benchmarkn/a

Retrieval

MTEB · Embeddings / retrieval · Retrieval

It is one of the few direct signals for retrieval stacks, where embedding quality matters more than chat style.

Rank #4 · Source label: gpt-5

backfilledproxy backfilledBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: MTEB
Raw value: 58.8 ndcg
Percentile: 100%
Last updated: archived
Eligibility: Fallback benchmark identity is visible for context but excluded from default ranking.

Embedding endpoint score. Backfilled from GPT-5 via approved benchmark identity mapping map-gpt-5-4-nano-to-gpt-5.

100% percentile inside its fair comparison set

58.8 ndcgRaw benchmark value

Multilingual20 benchmarksn/a

Text Arena · Chinese

AR · Multilingual · Human

Observed user preference in Arena's Text Arena chinese leaderboard.

Rank #150 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,422
Percentile: 57.7%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: chinese. Source rank: #167. Votes: 3826. Organization: openai. License: Proprietary.

57.7% percentile inside its fair comparison set

1,422Raw benchmark valueCI 1,412 - 1,433

Text Arena · French

AR · Multilingual · Human

Observed user preference in Arena's Text Arena french leaderboard.

Rank #109 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,429
Percentile: 58%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: french. Source rank: #122. Votes: 2098. Organization: openai. License: Proprietary.

58% percentile inside its fair comparison set

1,429Raw benchmark valueCI 1,414 - 1,445

Text Arena · German

AR · Multilingual · Human

Observed user preference in Arena's Text Arena german leaderboard.

Rank #109 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,394
Percentile: 60.3%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: german. Source rank: #125. Votes: 931. Organization: openai. License: Proprietary.

60.3% percentile inside its fair comparison set

1,394Raw benchmark valueCI 1,373 - 1,415

Text Arena · Japanese

AR · Multilingual · Human

Observed user preference in Arena's Text Arena japanese leaderboard.

Rank #92 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,373
Percentile: 62.6%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: japanese. Source rank: #107. Votes: 603. Organization: openai. License: Proprietary.

62.6% percentile inside its fair comparison set

1,373Raw benchmark valueCI 1,346 - 1,399

Text Arena · Korean

AR · Multilingual · Human

Observed user preference in Arena's Text Arena korean leaderboard.

Rank #124 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,344
Percentile: 49.6%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: korean. Source rank: #141. Votes: 1031. Organization: openai. License: Proprietary.

49.6% percentile inside its fair comparison set

1,344Raw benchmark valueCI 1,324 - 1,364

Text Arena · Non English

AR · Multilingual · Human

Observed user preference in Arena's Text Arena non_english leaderboard.

Rank #141 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,386
Percentile: 62.8%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: non_english. Source rank: #158. Votes: 33441. Organization: openai. License: Proprietary.

62.8% percentile inside its fair comparison set

1,386Raw benchmark valueCI 1,382 - 1,391

Text Arena · Polish

AR · Multilingual · Human

Observed user preference in Arena's Text Arena polish leaderboard.

Rank #107 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,402
Percentile: 46.7%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: polish. Source rank: #122. Votes: 1130. Organization: openai. License: Proprietary.

46.7% percentile inside its fair comparison set

1,402Raw benchmark valueCI 1,383 - 1,420

Text Arena · Russian

AR · Multilingual · Human

Observed user preference in Arena's Text Arena russian leaderboard.

Rank #143 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,389
Percentile: 58.4%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: russian. Source rank: #161. Votes: 6875. Organization: openai. License: Proprietary.

58.4% percentile inside its fair comparison set

1,389Raw benchmark valueCI 1,381 - 1,397

Text Arena · Spanish

AR · Multilingual · Human

Observed user preference in Arena's Text Arena spanish leaderboard.

Rank #136 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,393
Percentile: 47.9%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: spanish. Source rank: #154. Votes: 1912. Organization: openai. License: Proprietary.

47.9% percentile inside its fair comparison set

1,393Raw benchmark valueCI 1,378 - 1,408

Text Arena · Chinese · No Style Control

AR · Multilingual · Human

Observed user preference in Arena's Text Arena chinese leaderboard.

Rank #164 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,391
Percentile: 53.7%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: chinese. Source rank: #182. Votes: 3826. Organization: openai. License: Proprietary.

53.7% percentile inside its fair comparison set

1,391Raw benchmark valueCI 1,381 - 1,402

Text Arena · French · No Style Control

AR · Multilingual · Human

Observed user preference in Arena's Text Arena french leaderboard.

Rank #132 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,396
Percentile: 49%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: french. Source rank: #147. Votes: 2098. Organization: openai. License: Proprietary.

49% percentile inside its fair comparison set

1,396Raw benchmark valueCI 1,381 - 1,411

Text Arena · German · No Style Control

AR · Multilingual · Human

Observed user preference in Arena's Text Arena german leaderboard.

Rank #131 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,368
Percentile: 52.2%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: german. Source rank: #149. Votes: 931. Organization: openai. License: Proprietary.

52.2% percentile inside its fair comparison set

1,368Raw benchmark valueCI 1,347 - 1,389

Text Arena · Japanese · No Style Control

AR · Multilingual · Human

Observed user preference in Arena's Text Arena japanese leaderboard.

Rank #110 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,341
Percentile: 55.1%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: japanese. Source rank: #126. Votes: 603. Organization: openai. License: Proprietary.

55.1% percentile inside its fair comparison set

1,341Raw benchmark valueCI 1,315 - 1,367

Text Arena · Korean · No Style Control

AR · Multilingual · Human

Observed user preference in Arena's Text Arena korean leaderboard.

Rank #133 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,322
Percentile: 45.9%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: korean. Source rank: #150. Votes: 1031. Organization: openai. License: Proprietary.

45.9% percentile inside its fair comparison set

1,322Raw benchmark valueCI 1,302 - 1,341

Text Arena · Non English · No Style Control

AR · Multilingual · Human

Observed user preference in Arena's Text Arena non_english leaderboard.

Rank #158 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,359
Percentile: 58.2%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: non_english. Source rank: #177. Votes: 33441. Organization: openai. License: Proprietary.

58.2% percentile inside its fair comparison set

1,359Raw benchmark valueCI 1,355 - 1,364

Text Arena · Polish · No Style Control

AR · Multilingual · Human

Observed user preference in Arena's Text Arena polish leaderboard.

Rank #131 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,365
Percentile: 34.7%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: polish. Source rank: #149. Votes: 1130. Organization: openai. License: Proprietary.

34.7% percentile inside its fair comparison set

1,365Raw benchmark valueCI 1,347 - 1,384

Text Arena · Russian · No Style Control

AR · Multilingual · Human

Observed user preference in Arena's Text Arena russian leaderboard.

Rank #157 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,362
Percentile: 54.3%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: russian. Source rank: #176. Votes: 6875. Organization: openai. License: Proprietary.

54.3% percentile inside its fair comparison set

1,362Raw benchmark valueCI 1,354 - 1,370

Text Arena · Spanish · No Style Control

AR · Multilingual · Human

Observed user preference in Arena's Text Arena spanish leaderboard.

Rank #151 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,370
Percentile: 42.1%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: spanish. Source rank: #168. Votes: 1912. Organization: openai. License: Proprietary.

42.1% percentile inside its fair comparison set

1,370Raw benchmark valueCI 1,355 - 1,385

Vision Arena · Chinese

AR · Multilingual · Human

Observed user preference in Arena's Vision Arena chinese leaderboard.

Rank #61 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,236
Percentile: 42.3%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: chinese. Source rank: #70. Votes: 1734. Organization: openai. License: Proprietary.

42.3% percentile inside its fair comparison set

1,236Raw benchmark valueCI 1,219 - 1,254

Vision Arena · Chinese · No Style Control

AR · Multilingual · Human

Observed user preference in Arena's Vision Arena chinese leaderboard.

Rank #66 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,232
Percentile: 37.5%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: chinese. Source rank: #76. Votes: 1734. Organization: openai. License: Proprietary.

37.5% percentile inside its fair comparison set

1,232Raw benchmark valueCI 1,215 - 1,249

Source links and registry checks

official

Terminal-Bench

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Artificial Analysis

Oct 3, 2026

source →

official

LiveBench

Oct 3, 2026

source →

Model profile · OpenAI

GPT-5.4 nano

Closed weightsbudget · registry tag 2026 nano

Verified but aging

Reads as verified but aging across the resolved source data. Capability fit excludes speed and cost.

Visible coverage: 63.7%
Verified coverage: 56.2%
Spread: 83.1%
Last verified: Oct 3, 2026

33%capability fit

textcodevisiondocumentsearch18 aliases42 official source links

Open compare

Data version

Current snapshot.

Published Oct 3, 2026Model list checked4 stale sources105 claim warnings9 providers · 1334 tracked models

The registry snapshot and page stamp are shown so a stale deploy is visible at a glance. View live data status.

Source-linked scores by benchmark

Verified but agingThis model currently reads as verified but aging across the resolved source data.

Chat / text36 benchmarks35.5%

Intelligence Index

AA · Chat / text · Combined

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #405 · Source label: GPT-5 nano (Minimal)

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 7
Percentile: 27.2%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `intelligenceIndex`.

27.2% percentile inside its fair comparison set

7Raw benchmark value

AA-Omniscience accuracy

AA · Chat / text · Objective

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #363 · Source label: GPT-5 nano (Minimal)

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 13.1%
Percentile: 20.1%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `omniscienceAccuracy`.

20.1% percentile inside its fair comparison set

13.1%Raw benchmark value

AA-Omniscience non-hallucination

AA · Chat / text · Objective

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #313 · Source label: GPT-5 nano (Minimal)

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 11.2%
Percentile: 31.1%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `omniscienceNonHallucination`.

31.1% percentile inside its fair comparison set

11.2%Raw benchmark value

IFBench

AA · Chat / text · Objective

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #266 · Source label: GPT-5 nano (Minimal)

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 32.5%
Percentile: 23.9%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `ifbench`.

23.9% percentile inside its fair comparison set

32.5%Raw benchmark value

Input price

AA · Chat / text · Speed / cost

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #98 · Source label: GPT-5.4 nano (Xhigh)

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: $0.2 /1M input tokens
Percentile: 72.6%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `price1mInputTokens`.

72.6% percentile inside its fair comparison set

$0.2 /1M input tokensRaw benchmark value

Output price

AA · Chat / text · Speed / cost

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #148 · Source label: GPT-5.4 nano (Xhigh)

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: $1.3 /1M output tokens
Percentile: 58.5%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `price1mOutputTokens`.

58.5% percentile inside its fair comparison set

$1.3 /1M output tokensRaw benchmark value

Output Speed

AA · Chat / text · Speed / cost

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #57 · Source label: GPT-5 nano (Minimal)

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 143.2 tokens/s
Percentile: 78.6%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

78.6% percentile inside its fair comparison set

143.2 tokens/sRaw benchmark value

Time to first token

AA · Chat / text · Speed / cost

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #245 · Source label: GPT-5 nano (High)

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 89.60s
Percentile: 7.6%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `medianTimeToFirstTokenSeconds`.

7.6% percentile inside its fair comparison set

89.60sRaw benchmark value

Time to first answer token

AA · Chat / text · Speed / cost

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #239 · Source label: GPT-5 nano (High)

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 89.60s
Percentile: 9.2%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `medianTimeToFirstAnswerTokenSeconds`.

9.2% percentile inside its fair comparison set

89.60sRaw benchmark value

Text Arena

AR · Chat / text · Human

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #141 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,401
Percentile: 62.8%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: overall. Source rank: #158. Votes: 62249. Organization: openai. License: Proprietary.

62.8% percentile inside its fair comparison set

1,401Raw benchmark valueCI 1,398 - 1,405

Text Arena · Creative Writing

AR · Chat / text · Human

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #172 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,336
Percentile: 54.3%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: creative_writing. Source rank: #194. Votes: 10368. Organization: openai. License: Proprietary.

54.3% percentile inside its fair comparison set

1,336Raw benchmark valueCI 1,329 - 1,344

Text Arena · English

AR · Chat / text · Human

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #145 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,413
Percentile: 61.7%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: english. Source rank: #161. Votes: 28808. Organization: openai. License: Proprietary.

61.7% percentile inside its fair comparison set

1,413Raw benchmark valueCI 1,408 - 1,418

Text Arena · Exclude Ties

AR · Chat / text · Human

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #137 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,389
Percentile: 63.8%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: exclude_ties. Source rank: #153. Votes: 47863. Organization: openai. License: Proprietary.

63.8% percentile inside its fair comparison set

1,389Raw benchmark valueCI 1,383 - 1,394

Text Arena · Hard Prompts

AR · Chat / text · Human

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #138 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,421
Percentile: 63.6%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: hard_prompts. Source rank: #153. Votes: 41240. Organization: openai. License: Proprietary.

63.6% percentile inside its fair comparison set

1,421Raw benchmark valueCI 1,417 - 1,426

Text Arena · Hard Prompts English

AR · Chat / text · Human

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #142 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,429
Percentile: 62.3%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: hard_prompts_english. Source rank: #157. Votes: 19866. Organization: openai. License: Proprietary.

62.3% percentile inside its fair comparison set

1,429Raw benchmark valueCI 1,424 - 1,435

Text Arena · Instruction Following

AR · Chat / text · Human

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #143 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,387
Percentile: 62.2%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: instruction_following. Source rank: #161. Votes: 21407. Organization: openai. License: Proprietary.

62.2% percentile inside its fair comparison set

1,387Raw benchmark valueCI 1,381 - 1,392

Text Arena · Longer Query

AR · Chat / text · Human

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #147 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,403
Percentile: 58.9%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: longer_query. Source rank: #166. Votes: 27656. Organization: openai. License: Proprietary.

58.9% percentile inside its fair comparison set

1,403Raw benchmark valueCI 1,397 - 1,408

Text Arena · Multi Turn

AR · Chat / text · Human

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #129 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,413
Percentile: 65.8%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: multi_turn. Source rank: #144. Votes: 11793. Organization: openai. License: Proprietary.

65.8% percentile inside its fair comparison set

1,413Raw benchmark valueCI 1,406 - 1,420

Text Arena · No Style Control

AR · Chat / text · Human

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #161 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,372
Percentile: 57.4%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: overall. Source rank: #179. Votes: 62249. Organization: openai. License: Proprietary.

57.4% percentile inside its fair comparison set

1,372Raw benchmark valueCI 1,368 - 1,376

Text Arena · Creative Writing · No Style Control

AR · Chat / text · Human

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #185 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,314
Percentile: 50.8%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: creative_writing. Source rank: #208. Votes: 10368. Organization: openai. License: Proprietary.

50.8% percentile inside its fair comparison set

1,314Raw benchmark valueCI 1,307 - 1,321

Text Arena · English · No Style Control

AR · Chat / text · Human

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #167 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,381
Percentile: 55.9%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: english. Source rank: #184. Votes: 28808. Organization: openai. License: Proprietary.

55.9% percentile inside its fair comparison set

1,381Raw benchmark valueCI 1,376 - 1,386

Text Arena · Exclude Ties · No Style Control

AR · Chat / text · Human

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #161 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,345
Percentile: 57.4%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: exclude_ties. Source rank: #178. Votes: 47863. Organization: openai. License: Proprietary.

57.4% percentile inside its fair comparison set

1,345Raw benchmark valueCI 1,340 - 1,350

Text Arena · Hard Prompts · No Style Control

AR · Chat / text · Human

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #155 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,381
Percentile: 59%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: hard_prompts. Source rank: #172. Votes: 41240. Organization: openai. License: Proprietary.

59% percentile inside its fair comparison set

1,381Raw benchmark valueCI 1,376 - 1,385

Text Arena · Hard Prompts English · No Style Control

AR · Chat / text · Human

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #158 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,387
Percentile: 58%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: hard_prompts_english. Source rank: #175. Votes: 19866. Organization: openai. License: Proprietary.

58% percentile inside its fair comparison set

1,387Raw benchmark valueCI 1,382 - 1,393

Text Arena · Instruction Following · No Style Control

AR · Chat / text · Human

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #155 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,362
Percentile: 59%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: instruction_following. Source rank: #173. Votes: 21407. Organization: openai. License: Proprietary.

59% percentile inside its fair comparison set

1,362Raw benchmark valueCI 1,356 - 1,367

Text Arena · Longer Query · No Style Control

AR · Chat / text · Human

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #161 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,366
Percentile: 54.9%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: longer_query. Source rank: #181. Votes: 27656. Organization: openai. License: Proprietary.

54.9% percentile inside its fair comparison set

1,366Raw benchmark valueCI 1,360 - 1,371

Text Arena · Multi Turn · No Style Control

AR · Chat / text · Human

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #153 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,382
Percentile: 59.4%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: multi_turn. Source rank: #170. Votes: 11793. Organization: openai. License: Proprietary.

59.4% percentile inside its fair comparison set

1,382Raw benchmark valueCI 1,375 - 1,389

Instruction following

LB · Chat / text · Objective

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #38 · Source label: gpt-5.4-nano-xhigh

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 67.2%
Percentile: 43.1%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Category: IF. Tasks scored: 4.

43.1% percentile inside its fair comparison set

67.2%Raw benchmark value

Language

LB · Chat / text · Objective

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #66 · Source label: gpt-5.4-nano-xhigh

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 62.5%
Percentile: 0%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Category: Language. Tasks scored: 3.

0% percentile inside its fair comparison set

62.5%Raw benchmark value

Paraphrase

LB · Chat / text · Objective

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #34 · Source label: gpt-5.4-nano-xhigh

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 66.6%
Percentile: 49.2%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Task: paraphrase. Category: IF.

49.2% percentile inside its fair comparison set

66.6%Raw benchmark value

Simplify

LB · Chat / text · Objective

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #38 · Source label: gpt-5.4-nano-xhigh

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 62.8%
Percentile: 43.1%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Task: simplify. Category: IF.

43.1% percentile inside its fair comparison set

62.8%Raw benchmark value

Story generation

LB · Chat / text · Objective

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #49 · Source label: gpt-5.4-nano-xhigh

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 67.7%
Percentile: 26.2%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Task: story_generation. Category: IF.

26.2% percentile inside its fair comparison set

67.7%Raw benchmark value

Summarize

LB · Chat / text · Objective

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #30 · Source label: gpt-5.4-nano-xhigh

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 71.8%
Percentile: 55.4%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Task: summarize. Category: IF.

55.4% percentile inside its fair comparison set

71.8%Raw benchmark value

Connections

LB · Chat / text · Objective

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #48 · Source label: gpt-5.4-nano-xhigh

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 94.4%
Percentile: 27.7%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Task: connections. Category: Language.

27.7% percentile inside its fair comparison set

94.4%Raw benchmark value

Plot unscrambling

LB · Chat / text · Objective

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #66 · Source label: gpt-5.4-nano-xhigh

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 39.1%
Percentile: 0%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Task: plot_unscrambling. Category: Language.

0% percentile inside its fair comparison set

39.1%Raw benchmark value

Typos

LB · Chat / text · Objective

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #66 · Source label: gpt-5.4-nano-xhigh

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 54%
Percentile: 0%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Task: typos. Category: Language.

0% percentile inside its fair comparison set

54%Raw benchmark value

Coding28 benchmarks38.5%

Terminal-Bench Hard

AA · Coding · Objective

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #179 · Source label: GPT-5 nano (Minimal)

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 6.8%
Percentile: 46.4%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `terminalbenchHard`.

46.4% percentile inside its fair comparison set

6.8%Raw benchmark value

SciCode

AA · Coding · Objective

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #101 · Source label: GPT-5.4 nano (Xhigh)

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 47.2%
Percentile: 49.5%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `scicode`.

49.5% percentile inside its fair comparison set

47.2%Raw benchmark value

Text Arena · Coding

AR · Coding · Human

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #121 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,460
Percentile: 67.7%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: coding. Source rank: #136. Votes: 17626. Organization: openai. License: Proprietary.

67.7% percentile inside its fair comparison set

1,460Raw benchmark valueCI 1,454 - 1,466

Text Arena · Coding · No Style Control

AR · Coding · Human

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #150 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,405
Percentile: 59.8%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: coding. Source rank: #166. Votes: 17626. Organization: openai. License: Proprietary.

59.8% percentile inside its fair comparison set

1,405Raw benchmark valueCI 1,399 - 1,410

Vibe Code Bench v1.1

VALS-AI · Coding · Objective

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #59 · Source label: openai/gpt-5.4-nano-2026-03-17

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 26.1%
Percentile: 37.6%
Last updated: recent
Eligibility: headline eligible

Parsed from Vals AI BenchmarkView overall scores. Vals slug: vibe-code; provider: OpenAI.

37.6% percentile inside its fair comparison set

26.1%Raw benchmark valueCI 16.2% - 36%

WebDev Arena

AR · Coding · Human

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #76 · Source label: gpt-5-medium

backfilledproxy backfilledBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,417
Percentile: 41.7%
Last updated: recent
Eligibility: Fallback benchmark identity is visible for context but excluded from default ranking.

41.7% percentile inside its fair comparison set

1,417Raw benchmark valueCI 1,402 - 1,433

Code Arena · Webdev Html

AR · Coding · Human

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #72 · Source label: gpt-5-medium

backfilledproxy backfilledBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,427
Percentile: 44%
Last updated: recent
Eligibility: Fallback benchmark identity is visible for context but excluded from default ranking.

44% percentile inside its fair comparison set

1,427Raw benchmark valueCI 1,410 - 1,443

Code Migration

VALS-AI · Coding · Objective

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #49 · Source label: openai/gpt-5.4-nano-2026-03-17

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 14.5%
Percentile: 29.4%
Last updated: recent
Eligibility: headline eligible

Parsed from Vals AI BenchmarkView overall scores. Vals slug: code-migration; provider: OpenAI.

29.4% percentile inside its fair comparison set

14.5%Raw benchmark valueCI 6.6% - 22.4%

Terminal-Bench 2.1

VALS-AI · Coding · Objective

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #62 · Source label: openai/gpt-5.4-nano-2026-03-17

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 41.6%
Percentile: 9%
Last updated: recent
Eligibility: headline eligible

Parsed from Vals AI BenchmarkView overall scores. Vals slug: terminal-bench-2-1; provider: OpenAI.

9% percentile inside its fair comparison set

41.6%Raw benchmark valueCI 35.7% - 47.4%

SWE-bench Verified

VALS-AI · Coding · Objective

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #60 · Source label: openai/gpt-5.4-nano-2026-03-17

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 69.8%
Percentile: 28%
Last updated: recent
Eligibility: headline eligible

Parsed from Vals AI BenchmarkView overall scores. Vals slug: swebench; provider: OpenAI.

28% percentile inside its fair comparison set

69.8%Raw benchmark valueCI 65.8% - 73.8%

LiveCodeBench

VALS-AI · Coding · Objective

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #40 · Source label: openai/gpt-5.4-nano-2026-03-17

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 84%
Percentile: 68.8%
Last updated: recent
Eligibility: headline eligible

Parsed from Vals AI BenchmarkView overall scores. Vals slug: lcb; provider: OpenAI.

68.8% percentile inside its fair comparison set

84%Raw benchmark valueCI 82% - 86.1%

Agentic coding

LB · Coding · Objective

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #48 · Source label: gpt-5.4-nano-xhigh

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 46.8%
Percentile: 27.7%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Category: Agentic Coding. Tasks scored: 3.

27.7% percentile inside its fair comparison set

46.8%Raw benchmark value

Coding

LB · Coding · Objective

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #58 · Source label: gpt-5.4-nano-xhigh

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 70.8%
Percentile: 12.3%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Category: Coding. Tasks scored: 2.

12.3% percentile inside its fair comparison set

70.8%Raw benchmark value

JavaScript

LB · Coding · Objective

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #32 · Source label: gpt-5.4-nano-xhigh

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 63.6%
Percentile: 58.5%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Task: javascript. Category: Agentic Coding.

58.5% percentile inside its fair comparison set

63.6%Raw benchmark value

TypeScript

LB · Coding · Objective

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #48 · Source label: gpt-5.4-nano-xhigh

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 26.7%
Percentile: 27.7%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Task: typescript. Category: Agentic Coding.

27.7% percentile inside its fair comparison set

26.7%Raw benchmark value

Python

LB · Coding · Objective

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #41 · Source label: gpt-5.4-nano-xhigh

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 50%
Percentile: 46.2%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Task: python. Category: Agentic Coding.

46.2% percentile inside its fair comparison set

50%Raw benchmark value

Coding generation

LB · Coding · Objective

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #51 · Source label: gpt-5.4-nano-xhigh

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 74.3%
Percentile: 23.1%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Task: code_generation. Category: Coding.

23.1% percentile inside its fair comparison set

74.3%Raw benchmark value

Coding completion

LB · Coding · Objective

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #60 · Source label: gpt-5.4-nano-xhigh

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 67.4%
Percentile: 9.2%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Task: code_completion. Category: Coding.

9.2% percentile inside its fair comparison set

67.4%Raw benchmark value

Debugging

BB · Coding · Rubric

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #15 · Source label: openai/gpt-5.4-nano

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: BridgeBench
Raw value: 81.2%
Percentile: 41.7%
Last updated: archived
Eligibility: headline eligible

Parsed from the BridgeBench page for bridgebench-debugging.

41.7% percentile inside its fair comparison set

81.2%Raw benchmark value

Refactoring

BB · Coding · Rubric

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #16 · Source label: openai/gpt-5.4-nano

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: BridgeBench
Raw value: 56.9%
Percentile: 28.6%
Last updated: archived
Eligibility: headline eligible

Parsed from the BridgeBench page for bridgebench-refactoring.

28.6% percentile inside its fair comparison set

56.9%Raw benchmark value

Security

BB · Coding · Rubric

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #12 · Source label: openai/gpt-5.4-nano

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: BridgeBench
Raw value: 80%
Percentile: 56%
Last updated: archived
Eligibility: headline eligible

Parsed from the BridgeBench page for bridgebench-security.

56% percentile inside its fair comparison set

80%Raw benchmark value

UI bench

BB · Coding · Rubric

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #13 · Source label: openai/gpt-5.4-nano

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: BridgeBench
Raw value: 65.3%
Percentile: 45.5%
Last updated: archived
Eligibility: headline eligible

Parsed from the BridgeBench page for bridgebench-ui.

45.5% percentile inside its fair comparison set

65.3%Raw benchmark value

Speed throughput

BB · Coding · Speed / cost

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #5 · Source label: openai/gpt-5.4-nano

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: BridgeBench
Raw value: 227.8 t/s
Percentile: 84%
Last updated: archived
Eligibility: headline eligible

Parsed from the BridgeBench page for bridgebench-throughput.

84% percentile inside its fair comparison set

227.8 t/sRaw benchmark value

Speed TTFT

BB · Coding · Speed / cost

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #7 · Source label: openai/gpt-5.4-nano

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: BridgeBench
Raw value: 941.00ms
Percentile: 76%
Last updated: archived
Eligibility: headline eligible

Parsed from the BridgeBench page for bridgebench-ttft.

76% percentile inside its fair comparison set

941.00msRaw benchmark value

Speed cost

BB · Coding · Speed / cost

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #5 · Source label: openai/gpt-5.4-nano

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: BridgeBench
Raw value: $0 $
Percentile: 84%
Last updated: archived
Eligibility: headline eligible

Parsed from the BridgeBench page for bridgebench-speed-cost.

84% percentile inside its fair comparison set

$0 $Raw benchmark value

Overall quality

BB · Coding · Combined

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #13 · Source label: GPT-5.4 Nano

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: BridgeBench
Raw value: 70.9%
Percentile: 45.5%
Last updated: archived
Eligibility: headline eligible

Parsed from the BridgeBench page for bridgebench-overall-quality.

45.5% percentile inside its fair comparison set

70.9%Raw benchmark value

Overall vibe

BB · Coding · Combined

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #13 · Source label: GPT-5.4 Nano

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: BridgeBench
Raw value: 71.6%
Percentile: 45.5%
Last updated: archived
Eligibility: headline eligible

Parsed from the BridgeBench page for bridgebench-overall-vibe.

45.5% percentile inside its fair comparison set

71.6%Raw benchmark value

Overall coverage

BB · Coding · Combined

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #20 · Source label: GPT-5.4 Nano

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: BridgeBench
Raw value: 85.7%
Percentile: 22.7%
Last updated: archived
Eligibility: headline eligible

Parsed from the BridgeBench page for bridgebench-overall-coverage.

22.7% percentile inside its fair comparison set

85.7%Raw benchmark value

Reasoning / math / science20 benchmarks28.9%

Humanity's Last Exam

AA · Reasoning / math / science · Objective

It is one of the cleaner reads on deliberate reasoning strength rather than style or popularity.

Rank #458 · Source label: GPT-5 nano (Minimal)

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 4%
Percentile: 13.6%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `hle`.

13.6% percentile inside its fair comparison set

4%Raw benchmark value

GPQA

AA · Reasoning / math / science · Objective

It is one of the cleaner reads on deliberate reasoning strength rather than style or popularity.

Rank #386 · Source label: GPT-5 nano (Minimal)

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 42.8%
Percentile: 21.9%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `gpqa`.

21.9% percentile inside its fair comparison set

42.8%Raw benchmark value

CritPt

AA · Reasoning / math / science · Objective

It is one of the cleaner reads on deliberate reasoning strength rather than style or popularity.

Rank #230 · Source label: GPT-5 nano (High)

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 0%
Percentile: 50.1%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `critpt`.

50.1% percentile inside its fair comparison set

0%Raw benchmark value

TutorBench

SL · Reasoning / math / science · Rubric

It is one of the cleaner reads on deliberate reasoning strength rather than style or popularity.

Rank #6 · Source label: gpt-5-2025-08-07

backfilledproxy backfilledBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Scale Labs
Raw value: 55.3%
Percentile: 88.9%
Last updated: recent
Eligibility: Fallback benchmark identity is visible for context but excluded from default ranking.

88.9% percentile inside its fair comparison set

55.3%Raw benchmark value

MultiNRC

SL · Reasoning / math / science · Rubric

It is one of the cleaner reads on deliberate reasoning strength rather than style or popularity.

Rank #4 · Source label: gpt-5-pro-2025-10-06

backfilledproxy backfilledBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Scale Labs
Raw value: 65.2%
Percentile: 97.7%
Last updated: recent
Eligibility: Fallback benchmark identity is visible for context but excluded from default ranking.

97.7% percentile inside its fair comparison set

65.2%Raw benchmark value

Text Arena · Math

AR · Reasoning / math / science · Human

It is one of the cleaner reads on deliberate reasoning strength rather than style or popularity.

Rank #106 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,425
Percentile: 70.8%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: math. Source rank: #117. Votes: 3257. Organization: openai. License: Proprietary.

70.8% percentile inside its fair comparison set

1,425Raw benchmark valueCI 1,414 - 1,436

Text Arena · Math · No Style Control

AR · Reasoning / math / science · Human

It is one of the cleaner reads on deliberate reasoning strength rather than style or popularity.

Rank #133 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,406
Percentile: 63.2%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: math. Source rank: #146. Votes: 3257. Organization: openai. License: Proprietary.

63.2% percentile inside its fair comparison set

1,406Raw benchmark valueCI 1,395 - 1,417

GPQA Diamond

VALS-AI · Reasoning / math / science · Objective

It is one of the cleaner reads on deliberate reasoning strength rather than style or popularity.

Rank #76 · Source label: openai/gpt-5.4-nano-2026-03-17

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 77.5%
Percentile: 39.5%
Last updated: recent
Eligibility: headline eligible

Parsed from Vals AI BenchmarkView overall scores. Vals slug: gpqa; provider: OpenAI.

39.5% percentile inside its fair comparison set

77.5%Raw benchmark valueCI 72.8% - 82.3%

MMLU Pro

VALS-AI · Reasoning / math / science · Objective

It is one of the cleaner reads on deliberate reasoning strength rather than style or popularity.

Rank #102 · Source label: openai/gpt-5.4-nano-2026-03-17

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 77.2%
Percentile: 19.2%
Last updated: recent
Eligibility: headline eligible

Parsed from Vals AI BenchmarkView overall scores. Vals slug: mmlu_pro; provider: OpenAI.

19.2% percentile inside its fair comparison set

77.2%Raw benchmark valueCI 76.3% - 78%

Mathematics

LB · Reasoning / math / science · Objective

It is one of the cleaner reads on deliberate reasoning strength rather than style or popularity.

Rank #30 · Source label: gpt-5.4-nano-xhigh

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 91%
Percentile: 55.4%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Category: Mathematics. Tasks scored: 4.

55.4% percentile inside its fair comparison set

91%Raw benchmark value

Reasoning

LB · Reasoning / math / science · Objective

It is one of the cleaner reads on deliberate reasoning strength rather than style or popularity.

Rank #49 · Source label: gpt-5.4-nano-xhigh

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 81.1%
Percentile: 26.2%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Category: Reasoning. Tasks scored: 4.

26.2% percentile inside its fair comparison set

81.1%Raw benchmark value

AMPS Hard

LB · Reasoning / math / science · Objective

It is one of the cleaner reads on deliberate reasoning strength rather than style or popularity.

Rank #20 · Source label: gpt-5.4-nano-xhigh

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 98%
Percentile: 83.1%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Task: AMPS_Hard. Category: Mathematics.

83.1% percentile inside its fair comparison set

98%Raw benchmark value

Integrals with game

LB · Reasoning / math / science · Objective

It is one of the cleaner reads on deliberate reasoning strength rather than style or popularity.

Rank #29 · Source label: gpt-5.4-nano-xhigh

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 82.8%
Percentile: 56.9%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Task: integrals_with_game. Category: Mathematics.

56.9% percentile inside its fair comparison set

82.8%Raw benchmark value

Math competition

LB · Reasoning / math / science · Objective

It is one of the cleaner reads on deliberate reasoning strength rather than style or popularity.

Rank #40 · Source label: gpt-5.4-nano-xhigh

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 95.1%
Percentile: 49.2%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Task: math_comp. Category: Mathematics.

49.2% percentile inside its fair comparison set

95.1%Raw benchmark value

Olympiad

LB · Reasoning / math / science · Objective

It is one of the cleaner reads on deliberate reasoning strength rather than style or popularity.

Rank #48 · Source label: gpt-5.4-nano-xhigh

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 88%
Percentile: 27.7%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Task: olympiad. Category: Mathematics.

27.7% percentile inside its fair comparison set

88%Raw benchmark value

Theory of mind

LB · Reasoning / math / science · Objective

It is one of the cleaner reads on deliberate reasoning strength rather than style or popularity.

Rank #28 · Source label: gpt-5.4-nano-xhigh

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 80.8%
Percentile: 66.2%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Task: theory_of_mind. Category: Reasoning.

66.2% percentile inside its fair comparison set

80.8%Raw benchmark value

Zebra puzzle

LB · Reasoning / math / science · Objective

It is one of the cleaner reads on deliberate reasoning strength rather than style or popularity.

Rank #28 · Source label: gpt-5.4-nano-xhigh

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 97.6%
Percentile: 58.5%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Task: zebra_puzzle. Category: Reasoning.

58.5% percentile inside its fair comparison set

97.6%Raw benchmark value

Spatial

LB · Reasoning / math / science · Objective

It is one of the cleaner reads on deliberate reasoning strength rather than style or popularity.

Rank #66 · Source label: gpt-5.4-nano-xhigh

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 82%
Percentile: 0%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Task: spatial. Category: Reasoning.

0% percentile inside its fair comparison set

82%Raw benchmark value

Logic with navigation

LB · Reasoning / math / science · Objective

It is one of the cleaner reads on deliberate reasoning strength rather than style or popularity.

Rank #61 · Source label: gpt-5.4-nano-xhigh

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 64%
Percentile: 10.8%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Task: logic_with_navigation. Category: Reasoning.

10.8% percentile inside its fair comparison set

64%Raw benchmark value

Reasoning

BB · Reasoning / math / science · Rubric

It is one of the cleaner reads on deliberate reasoning strength rather than style or popularity.

Rank #19 · Source label: openai/gpt-5.4-nano

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: BridgeBench
Raw value: 34.8%
Percentile: 14.3%
Last updated: archived
Eligibility: headline eligible

Parsed from the BridgeBench page for bridgebench-reasoning.

14.3% percentile inside its fair comparison set

34.8%Raw benchmark value

Professional reasoning35 benchmarks36%

APEX-Agents-AA

AA · Professional reasoning · Objective

Long-horizon agentic task completion.

Rank #18 · Source label: GPT-5.4 nano (Xhigh)

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 24.9%
Percentile: 46.9%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `apexAgents`.

46.9% percentile inside its fair comparison set

24.9%Raw benchmark value

PRBench Legal

SL · Professional reasoning · Rubric

Applied legal reasoning on professional-domain tasks.

Rank #10 · Source label: gpt-5-pro

backfilledproxy backfilledBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Scale Labs
Raw value: 49.9%
Percentile: 82.9%
Last updated: recent
Eligibility: Fallback benchmark identity is visible for context but excluded from default ranking.

82.9% percentile inside its fair comparison set

49.9%Raw benchmark value

Text Arena · Expert

AR · Professional reasoning · Human

Observed user preference in Arena's Text Arena expert leaderboard.

Rank #121 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,437
Percentile: 63.3%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: expert. Source rank: #134. Votes: 6344. Organization: openai. License: Proprietary.

63.3% percentile inside its fair comparison set

1,437Raw benchmark valueCI 1,429 - 1,446

Text Arena · Industry Business And Management And Financial Operations

AR · Professional reasoning · Human

Observed user preference in Arena's Text Arena industry_business_and_management_and_financial_operations leaderboard.

Rank #137 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,405
Percentile: 63.1%
Last updated: recent
Eligibility: headline eligible

63.1% percentile inside its fair comparison set

1,405Raw benchmark valueCI 1,399 - 1,412

Text Arena · Industry Entertainment And Sports And Media

AR · Professional reasoning · Human

Observed user preference in Arena's Text Arena industry_entertainment_and_sports_and_media leaderboard.

Rank #153 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,352
Percentile: 59.4%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: industry_entertainment_and_sports_and_media. Source rank: #172. Votes: 13657. Organization: openai. License: Proprietary.

59.4% percentile inside its fair comparison set

1,352Raw benchmark valueCI 1,345 - 1,358

Text Arena · Industry Legal And Government

AR · Professional reasoning · Human

Observed user preference in Arena's Text Arena industry_legal_and_government leaderboard.

Rank #150 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,400
Percentile: 57.3%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: industry_legal_and_government. Source rank: #168. Votes: 4954. Organization: openai. License: Proprietary.

57.3% percentile inside its fair comparison set

1,400Raw benchmark valueCI 1,391 - 1,409

Text Arena · Industry Life And Physical And Social Science

AR · Professional reasoning · Human

Observed user preference in Arena's Text Arena industry_life_and_physical_and_social_science leaderboard.

Rank #145 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,415
Percentile: 61.5%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: industry_life_and_physical_and_social_science. Source rank: #162. Votes: 10492. Organization: openai. License: Proprietary.

61.5% percentile inside its fair comparison set

1,415Raw benchmark valueCI 1,408 - 1,422

Text Arena · Industry Mathematical

AR · Professional reasoning · Human

Observed user preference in Arena's Text Arena industry_mathematical leaderboard.

Rank #109 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,431
Percentile: 69.7%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: industry_mathematical. Source rank: #119. Votes: 3478. Organization: openai. License: Proprietary.

69.7% percentile inside its fair comparison set

1,431Raw benchmark valueCI 1,420 - 1,442

Text Arena · Industry Medicine And Healthcare

AR · Professional reasoning · Human

Observed user preference in Arena's Text Arena industry_medicine_and_healthcare leaderboard.

Rank #150 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,413
Percentile: 56.9%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: industry_medicine_and_healthcare. Source rank: #168. Votes: 4788. Organization: openai. License: Proprietary.

56.9% percentile inside its fair comparison set

1,413Raw benchmark valueCI 1,403 - 1,423

Text Arena · Industry Software And It Services

AR · Professional reasoning · Human

Observed user preference in Arena's Text Arena industry_software_and_it_services leaderboard.

Rank #125 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,450
Percentile: 67%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: industry_software_and_it_services. Source rank: #139. Votes: 24913. Organization: openai. License: Proprietary.

67% percentile inside its fair comparison set

1,450Raw benchmark valueCI 1,445 - 1,455

Text Arena · Industry Writing And Literature And Language

AR · Professional reasoning · Human

Observed user preference in Arena's Text Arena industry_writing_and_literature_and_language leaderboard.

Rank #159 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,363
Percentile: 57.9%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: industry_writing_and_literature_and_language. Source rank: #179. Votes: 15381. Organization: openai. License: Proprietary.

57.9% percentile inside its fair comparison set

1,363Raw benchmark valueCI 1,357 - 1,369

Text Arena · Expert · No Style Control

AR · Professional reasoning · Human

Observed user preference in Arena's Text Arena expert leaderboard.

Rank #142 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,396
Percentile: 56.9%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: expert. Source rank: #159. Votes: 6344. Organization: openai. License: Proprietary.

56.9% percentile inside its fair comparison set

1,396Raw benchmark valueCI 1,388 - 1,405

Text Arena · Industry Business And Management And Financial Operations · No Style Control

AR · Professional reasoning · Human

Observed user preference in Arena's Text Arena industry_business_and_management_and_financial_operations leaderboard.

Rank #158 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,366
Percentile: 57.5%
Last updated: recent
Eligibility: headline eligible

57.5% percentile inside its fair comparison set

1,366Raw benchmark valueCI 1,360 - 1,373

Text Arena · Industry Entertainment And Sports And Media · No Style Control

AR · Professional reasoning · Human

Observed user preference in Arena's Text Arena industry_entertainment_and_sports_and_media leaderboard.

Rank #172 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,324
Percentile: 54.3%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: industry_entertainment_and_sports_and_media. Source rank: #193. Votes: 13657. Organization: openai. License: Proprietary.

54.3% percentile inside its fair comparison set

1,324Raw benchmark valueCI 1,317 - 1,331

Text Arena · Industry Legal And Government · No Style Control

AR · Professional reasoning · Human

Observed user preference in Arena's Text Arena industry_legal_and_government leaderboard.

Rank #167 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,367
Percentile: 52.4%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: industry_legal_and_government. Source rank: #187. Votes: 4954. Organization: openai. License: Proprietary.

52.4% percentile inside its fair comparison set

1,367Raw benchmark valueCI 1,358 - 1,377

Text Arena · Industry Life And Physical And Social Science · No Style Control

AR · Professional reasoning · Human

Observed user preference in Arena's Text Arena industry_life_and_physical_and_social_science leaderboard.

Rank #170 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,374
Percentile: 54.8%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: industry_life_and_physical_and_social_science. Source rank: #187. Votes: 10492. Organization: openai. License: Proprietary.

54.8% percentile inside its fair comparison set

1,374Raw benchmark valueCI 1,367 - 1,381

Text Arena · Industry Mathematical · No Style Control

AR · Professional reasoning · Human

Observed user preference in Arena's Text Arena industry_mathematical leaderboard.

Rank #138 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,408
Percentile: 61.5%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: industry_mathematical. Source rank: #150. Votes: 3478. Organization: openai. License: Proprietary.

61.5% percentile inside its fair comparison set

1,408Raw benchmark valueCI 1,398 - 1,419

Text Arena · Industry Medicine And Healthcare · No Style Control

AR · Professional reasoning · Human

Observed user preference in Arena's Text Arena industry_medicine_and_healthcare leaderboard.

Rank #164 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,369
Percentile: 52.9%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: industry_medicine_and_healthcare. Source rank: #183. Votes: 4788. Organization: openai. License: Proprietary.

52.9% percentile inside its fair comparison set

1,369Raw benchmark valueCI 1,359 - 1,378

Text Arena · Industry Software And It Services · No Style Control

AR · Professional reasoning · Human

Observed user preference in Arena's Text Arena industry_software_and_it_services leaderboard.

Rank #154 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,402
Percentile: 59.3%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: industry_software_and_it_services. Source rank: #171. Votes: 24913. Organization: openai. License: Proprietary.

59.3% percentile inside its fair comparison set

1,402Raw benchmark valueCI 1,397 - 1,408

Text Arena · Industry Writing And Literature And Language · No Style Control

AR · Professional reasoning · Human

Observed user preference in Arena's Text Arena industry_writing_and_literature_and_language leaderboard.

Rank #170 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,338
Percentile: 54.9%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: industry_writing_and_literature_and_language. Source rank: #192. Votes: 15381. Organization: openai. License: Proprietary.

54.9% percentile inside its fair comparison set

1,338Raw benchmark valueCI 1,332 - 1,344

Legal Research Bench

VALS-AI · Professional reasoning · Objective

Applied legal research tasks.

Rank #64 · Source label: openai/gpt-5.4-nano-2026-03-17

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 6.3%
Percentile: 7.4%
Last updated: recent
Eligibility: headline eligible

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_research; provider: OpenAI.

7.4% percentile inside its fair comparison set

6.3%Raw benchmark valueCI 3% - 9.5%

Harvey's Legal Agent Benchmark

VALS-AI · Professional reasoning · Objective

Completing legal work with documents, spreadsheets, presentations, and file-system tools.

Rank #60 · Source label: openai/gpt-5.4-nano-2026-03-17

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 0%
Percentile: 20.3%
Last updated: recent
Eligibility: headline eligible

Parsed from Vals AI BenchmarkView overall scores. Vals slug: hlab; provider: OpenAI.

20.3% percentile inside its fair comparison set

0%Raw benchmark valueCI 0% - 0%

LegalBench

VALS-AI · Professional reasoning · Objective

Academic legal reasoning tasks.

Rank #97 · Source label: openai/gpt-5.4-nano-2026-03-17

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 77.9%
Percentile: 28.9%
Last updated: recent
Eligibility: headline eligible

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: OpenAI.

28.9% percentile inside its fair comparison set

77.9%Raw benchmark valueCI 77.1% - 78.8%

Finance Agent v2

VALS-AI · Professional reasoning · Objective

Core financial analyst tasks for agentic models.

Rank #56 · Source label: openai/gpt-5.4-nano-2026-03-17

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 38.2%
Percentile: 20.3%
Last updated: recent
Eligibility: headline eligible

Parsed from Vals AI BenchmarkView overall scores. Vals slug: fabv2; provider: OpenAI.

20.3% percentile inside its fair comparison set

38.2%Raw benchmark valueCI 35.9% - 40.5%

MedCode

VALS-AI · Professional reasoning · Objective

Medical billing support and coding tasks.

Rank #58 · Source label: openai/gpt-5.4-nano-2026-03-17

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 41%
Percentile: 40%
Last updated: recent
Eligibility: headline eligible

Parsed from Vals AI BenchmarkView overall scores. Vals slug: medcode; provider: OpenAI.

40% percentile inside its fair comparison set

41%Raw benchmark valueCI 36.6% - 45.5%

MedScribe

VALS-AI · Professional reasoning · Objective

Administrative documentation support for doctors.

Rank #64 · Source label: openai/gpt-5.4-nano-2026-03-17

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 77.1%
Percentile: 34.4%
Last updated: recent
Eligibility: headline eligible

Parsed from Vals AI BenchmarkView overall scores. Vals slug: medscribe; provider: OpenAI.

34.4% percentile inside its fair comparison set

77.1%Raw benchmark valueCI 73.4% - 80.8%

SAGE

VALS-AI · Professional reasoning · Objective

Student Assessment with Generative Evaluation.

Rank #60 · Source label: openai/gpt-5.4-nano-2026-03-17

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 38.1%
Percentile: 27.2%
Last updated: recent
Eligibility: headline eligible

Parsed from Vals AI BenchmarkView overall scores. Vals slug: sage; provider: OpenAI.

27.2% percentile inside its fair comparison set

38.1%Raw benchmark valueCI 32% - 44.1%

TaxEval v2

VALS-AI · Professional reasoning · Objective

Answer quality on tax questions and responses.

Rank #100 · Source label: openai/gpt-5.4-nano-2026-03-17

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 67.4%
Percentile: 23.3%
Last updated: recent
Eligibility: headline eligible

Parsed from Vals AI BenchmarkView overall scores. Vals slug: tax_eval_v2; provider: OpenAI.

23.3% percentile inside its fair comparison set

67.4%Raw benchmark valueCI 65.6% - 69.2%

Data analysis

LB · Professional reasoning · Objective

Structured data manipulation and table reasoning accuracy.

Rank #56 · Source label: gpt-5.4-nano-xhigh

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 67.6%
Percentile: 15.4%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Category: Data Analysis. Tasks scored: 3.

15.4% percentile inside its fair comparison set

67.6%Raw benchmark value

Overall

LB · Professional reasoning · Objective

Average objective performance across LiveBench's current public category mix.

Rank #55 · Source label: gpt-5.4-nano-xhigh

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 69.6%
Percentile: 16.9%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Category averages included: 7.

16.9% percentile inside its fair comparison set

69.6%Raw benchmark value

Consecutive events

LB · Professional reasoning · Objective

Objective consecutive events score in LiveBench.

Rank #56 · Source label: gpt-5.4-nano-xhigh

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 54.3%
Percentile: 15.4%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Task: consecutive_events. Category: Data Analysis.

15.4% percentile inside its fair comparison set

54.3%Raw benchmark value

Table join

LB · Professional reasoning · Objective

Objective table join score in LiveBench.

Rank #18 · Source label: gpt-5.4-nano-xhigh

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 50.6%
Percentile: 73.8%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Task: tablejoin. Category: Data Analysis.

73.8% percentile inside its fair comparison set

50.6%Raw benchmark value

Table reformat

LB · Professional reasoning · Objective

Objective table reformat score in LiveBench.

Rank #32 · Source label: gpt-5.4-nano-xhigh

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 98%
Percentile: 61.5%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Task: tablereformat. Category: Data Analysis.

61.5% percentile inside its fair comparison set

98%Raw benchmark value

Hallucination

BB · Professional reasoning · Rubric

Accuracy and fabrication resistance under prompts that invite unsupported claims.

Rank #23 · Source label: openai/gpt-5.4-nano

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: BridgeBench
Raw value: 69.6%
Percentile: 35.3%
Last updated: archived
Eligibility: headline eligible

Parsed from the BridgeBench page for bridgebench-hallucination.

35.3% percentile inside its fair comparison set

69.6%Raw benchmark value

Poker Agent

VALS-AI · Professional reasoning · Objective

Agent profit in poker-style strategic play.

Rank #5 · Source label: openai/gpt-5-2025-08-07

backfilledproxy backfilledBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 1,103.2 score
Percentile: 94.7%
Last updated: archived
Eligibility: Fallback benchmark identity is visible for context but excluded from default ranking.

Parsed from Vals AI BenchmarkView overall scores. Vals slug: poker_agent; provider: unknown. Backfilled from GPT-5 via approved benchmark identity mapping map-gpt-5-4-nano-to-gpt-5.

94.7% percentile inside its fair comparison set

1,103.2 scoreRaw benchmark valueCI 1,103.2 score - 1,103.2 score

Search / tool use3 benchmarks37.2%

Tau2-Bench Telecom

AA · Search / tool use · Objective

It matters when the model must browse, call tools, and recover useful answers from external systems.

Rank #214 · Source label: GPT-5 nano (Minimal)

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 25.7%
Percentile: 37.2%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `tau2`.

37.2% percentile inside its fair comparison set

25.7%Raw benchmark value

Search Arena

AR · Search / tool use · Human

It matters when the model must browse, call tools, and recover useful answers from external systems.

Rank #23 · Source label: gpt-5-search

backfilledproxy backfilledBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,183
Percentile: 39.4%
Last updated: recent
Eligibility: Fallback benchmark identity is visible for context but excluded from default ranking.

39.4% percentile inside its fair comparison set

1,183Raw benchmark valueCI 1,175 - 1,190

Search Arena · No Style Control

AR · Search / tool use · Human

It matters when the model must browse, call tools, and recover useful answers from external systems.

Rank #30 · Source label: gpt-5-search

backfilledproxy backfilledBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,133
Percentile: 18.2%
Last updated: recent
Eligibility: Fallback benchmark identity is visible for context but excluded from default ranking.

18.2% percentile inside its fair comparison set

1,133Raw benchmark valueCI 1,127 - 1,139

Long context2 benchmarks38.2%

Long Context Reasoning

AA · Long context · Objective

It checks whether long-context claims survive contact with retrieval, memory, or long-document tasks.

Rank #342 · Source label: GPT-5 nano (Minimal)

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 20%
Percentile: 25.1%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `lcr`.

25.1% percentile inside its fair comparison set

20%Raw benchmark value

CorpFin v2

VALS-AI · Long context · Objective

It checks whether long-context claims survive contact with retrieval, memory, or long-document tasks.

Rank #58 · Source label: openai/gpt-5.4-nano-2026-03-17

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 61.2%
Percentile: 51.3%
Last updated: aging
Eligibility: headline eligible

Parsed from Vals AI BenchmarkView overall scores. Vals slug: corp_fin_v2; provider: OpenAI.

51.3% percentile inside its fair comparison set

61.2%Raw benchmark valueCI 59.3% - 63.1%

Vision understanding24 benchmarks26.2%

MMMU-Pro

AA · Vision understanding · Objective

It is useful when the model must read charts, UI, screenshots, or visual scenes rather than text alone.

Rank #201 · Source label: GPT-5 nano (Minimal)

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 31.8%
Percentile: 3.8%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

3.8% percentile inside its fair comparison set

31.8%Raw benchmark value

VISTA

SL · Vision understanding · Rubric

It is useful when the model must read charts, UI, screenshots, or visual scenes rather than text alone.

Rank #6 · Source label: gpt-5-pro-2025-10-06

backfilledproxy backfilledBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Scale Labs
Raw value: 52.4%
Percentile: 95.2%
Last updated: recent
Eligibility: Fallback benchmark identity is visible for context but excluded from default ranking.

95.2% percentile inside its fair comparison set

52.4%Raw benchmark value

VTB

SL · Vision understanding · Rubric

It is useful when the model must read charts, UI, screenshots, or visual scenes rather than text alone.

Rank #8 · Source label: gpt-5-2025-08-07-thinking

backfilledproxy backfilledBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Scale Labs
Raw value: 18.7%
Percentile: 79.2%
Last updated: recent
Eligibility: Fallback benchmark identity is visible for context but excluded from default ranking.

79.2% percentile inside its fair comparison set

18.7%Raw benchmark value

Vision Arena

AR · Vision understanding · Human

It is useful when the model must read charts, UI, screenshots, or visual scenes rather than text alone.

Rank #78 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,199
Percentile: 44.6%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: overall. Source rank: #87. Votes: 25647. Organization: openai. License: Proprietary.

44.6% percentile inside its fair comparison set

1,199Raw benchmark valueCI 1,193 - 1,206

Vision Arena · Captioning

AR · Vision understanding · Human

It is useful when the model must read charts, UI, screenshots, or visual scenes rather than text alone.

Rank #16 · Source label: gpt-5-chat

backfilledproxy backfilledBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,195
Percentile: 58.1%
Last updated: recent
Eligibility: Fallback benchmark identity is visible for context but excluded from default ranking.

58.1% percentile inside its fair comparison set

1,195Raw benchmark valueCI 1,163 - 1,227

Vision Arena · Creative Writing Vision

AR · Vision understanding · Human

It is useful when the model must read charts, UI, screenshots, or visual scenes rather than text alone.

Rank #77 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,156
Percentile: 7.3%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: creative_writing_vision. Source rank: #88. Votes: 1530. Organization: openai. License: Proprietary.

7.3% percentile inside its fair comparison set

1,156Raw benchmark valueCI 1,140 - 1,173

Vision Arena · Diagram

AR · Vision understanding · Human

It is useful when the model must read charts, UI, screenshots, or visual scenes rather than text alone.

Rank #73 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,227
Percentile: 27.3%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: diagram. Source rank: #82. Votes: 6845. Organization: openai. License: Proprietary.

27.3% percentile inside its fair comparison set

1,227Raw benchmark valueCI 1,218 - 1,236

Vision Arena · English

AR · Vision understanding · Human

It is useful when the model must read charts, UI, screenshots, or visual scenes rather than text alone.

Rank #74 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,204
Percentile: 47.1%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: english. Source rank: #84. Votes: 10343. Organization: openai. License: Proprietary.

47.1% percentile inside its fair comparison set

1,204Raw benchmark valueCI 1,195 - 1,212

Vision Arena · Entity Recognition

AR · Vision understanding · Human

It is useful when the model must read charts, UI, screenshots, or visual scenes rather than text alone.

Rank #41 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,119
Percentile: 0%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: entity_recognition. Source rank: #50. Votes: 175. Organization: openai. License: Proprietary.

0% percentile inside its fair comparison set

1,119Raw benchmark valueCI 1,077 - 1,162

Vision Arena · Homework

AR · Vision understanding · Human

It is useful when the model must read charts, UI, screenshots, or visual scenes rather than text alone.

Rank #70 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,233
Percentile: 28.1%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: homework. Source rank: #81. Votes: 3447. Organization: openai. License: Proprietary.

28.1% percentile inside its fair comparison set

1,233Raw benchmark valueCI 1,222 - 1,244

Vision Arena · Humor

AR · Vision understanding · Human

It is useful when the model must read charts, UI, screenshots, or visual scenes rather than text alone.

Rank #67 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,155
Percentile: 9.6%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: humor. Source rank: #78. Votes: 922. Organization: openai. License: Proprietary.

9.6% percentile inside its fair comparison set

1,155Raw benchmark valueCI 1,134 - 1,176

Vision Arena · Ocr

AR · Vision understanding · Human

It is useful when the model must read charts, UI, screenshots, or visual scenes rather than text alone.

Rank #76 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,213
Percentile: 25%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: ocr. Source rank: #85. Votes: 18243. Organization: openai. License: Proprietary.

25% percentile inside its fair comparison set

1,213Raw benchmark valueCI 1,207 - 1,220

Vision Arena · No Style Control

AR · Vision understanding · Human

It is useful when the model must read charts, UI, screenshots, or visual scenes rather than text alone.

Rank #77 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,196
Percentile: 45.3%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: overall. Source rank: #88. Votes: 25647. Organization: openai. License: Proprietary.

45.3% percentile inside its fair comparison set

1,196Raw benchmark valueCI 1,190 - 1,202

Vision Arena · Captioning · No Style Control

AR · Vision understanding · Human

It is useful when the model must read charts, UI, screenshots, or visual scenes rather than text alone.

Rank #20 · Source label: gpt-5-chat

backfilledproxy backfilledBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,212
Percentile: 45.2%
Last updated: recent
Eligibility: Fallback benchmark identity is visible for context but excluded from default ranking.

45.2% percentile inside its fair comparison set

1,212Raw benchmark valueCI 1,181 - 1,244

Vision Arena · Creative Writing Vision · No Style Control

AR · Vision understanding · Human

It is useful when the model must read charts, UI, screenshots, or visual scenes rather than text alone.

Rank #77 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,157
Percentile: 7.3%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: creative_writing_vision. Source rank: #89. Votes: 1530. Organization: openai. License: Proprietary.

7.3% percentile inside its fair comparison set

1,157Raw benchmark valueCI 1,141 - 1,174

Vision Arena · Diagram · No Style Control

AR · Vision understanding · Human

It is useful when the model must read charts, UI, screenshots, or visual scenes rather than text alone.

Rank #73 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,211
Percentile: 27.3%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: diagram. Source rank: #83. Votes: 6845. Organization: openai. License: Proprietary.

27.3% percentile inside its fair comparison set

1,211Raw benchmark valueCI 1,202 - 1,220

Vision Arena · English · No Style Control

AR · Vision understanding · Human

It is useful when the model must read charts, UI, screenshots, or visual scenes rather than text alone.

Rank #76 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,201
Percentile: 45.7%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: english. Source rank: #87. Votes: 10343. Organization: openai. License: Proprietary.

45.7% percentile inside its fair comparison set

1,201Raw benchmark valueCI 1,193 - 1,210

Vision Arena · Entity Recognition · No Style Control

AR · Vision understanding · Human

It is useful when the model must read charts, UI, screenshots, or visual scenes rather than text alone.

Rank #40 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,143
Percentile: 2.5%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: entity_recognition. Source rank: #49. Votes: 175. Organization: openai. License: Proprietary.

2.5% percentile inside its fair comparison set

1,143Raw benchmark valueCI 1,101 - 1,186

Vision Arena · Homework · No Style Control

AR · Vision understanding · Human

It is useful when the model must read charts, UI, screenshots, or visual scenes rather than text alone.

Rank #73 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,225
Percentile: 25%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: homework. Source rank: #83. Votes: 3447. Organization: openai. License: Proprietary.

25% percentile inside its fair comparison set

1,225Raw benchmark valueCI 1,214 - 1,236

Vision Arena · Humor · No Style Control

AR · Vision understanding · Human

It is useful when the model must read charts, UI, screenshots, or visual scenes rather than text alone.

Rank #69 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,160
Percentile: 6.8%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: humor. Source rank: #81. Votes: 922. Organization: openai. License: Proprietary.

6.8% percentile inside its fair comparison set

1,160Raw benchmark valueCI 1,139 - 1,181

Vision Arena · Ocr · No Style Control

AR · Vision understanding · Human

It is useful when the model must read charts, UI, screenshots, or visual scenes rather than text alone.

Rank #74 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,204
Percentile: 27%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: ocr. Source rank: #84. Votes: 18243. Organization: openai. License: Proprietary.

27% percentile inside its fair comparison set

1,204Raw benchmark valueCI 1,198 - 1,211

MMMU Pro

VALS-AI · Vision understanding · Objective

It is useful when the model must read charts, UI, screenshots, or visual scenes rather than text alone.

Rank #59 · Source label: openai/gpt-5.4-nano-2026-03-17

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 73.6%
Percentile: 30.1%
Last updated: recent
Eligibility: headline eligible

Parsed from Vals AI BenchmarkView overall scores. Vals slug: mmmu; provider: OpenAI.

30.1% percentile inside its fair comparison set

73.6%Raw benchmark valueCI 71.5% - 75.7%

Vision Arena · Creative Writing

AR · Vision understanding · Human

It is useful when the model must read charts, UI, screenshots, or visual scenes rather than text alone.

Rank #33 · Source label: gpt-5-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,092
Percentile: 0%
Last updated: archived
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5-nano-high`. Category: creative_writing. Source rank: #34. Votes: 206. Organization: openai. License: Proprietary.

0% percentile inside its fair comparison set

1,092Raw benchmark valueCI 1,050 - 1,134

Vision Arena · Creative Writing · No Style Control

AR · Vision understanding · Human

It is useful when the model must read charts, UI, screenshots, or visual scenes rather than text alone.

Rank #32 · Source label: gpt-5-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,117
Percentile: 3.1%
Last updated: archived
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5-nano-high`. Category: creative_writing. Source rank: #33. Votes: 206. Organization: openai. License: Proprietary.

3.1% percentile inside its fair comparison set

1,117Raw benchmark valueCI 1,077 - 1,158

Document understanding3 benchmarks20%

MortgageTax

VALS-AI · Document understanding · Objective

It matters when the job is reading PDFs, tables, forms, or mixed-layout documents rather than plain chat.

Rank #66 · Source label: openai/gpt-5.4-nano-2026-03-17

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 59.1%
Percentile: 24.4%
Last updated: recent
Eligibility: headline eligible

Parsed from Vals AI BenchmarkView overall scores. Vals slug: mortgage_tax; provider: OpenAI.

24.4% percentile inside its fair comparison set

59.1%Raw benchmark valueCI 57.2% - 61%

Vals Multimodal Index

VALS-AI · Document understanding · Combined

It matters when the job is reading PDFs, tables, forms, or mixed-layout documents rather than plain chat.

Rank #28 · Source label: openai/gpt-5.4-nano-2026-03-17

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 48
Percentile: 15.6%
Last updated: aging
Eligibility: headline eligible

Parsed from Vals AI BenchmarkView overall scores. Vals slug: vals_multimodal_index; provider: OpenAI.

15.6% percentile inside its fair comparison set

48Raw benchmark valueCI 44 - 51

Multimodal mix

OC · Document understanding · Objective

It matters when the job is reading PDFs, tables, forms, or mixed-layout documents rather than plain chat.

Rank #5 · Source label: gpt-5

backfilledproxy backfilledBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: OpenCompass
Raw value: 75.4%
Percentile: 92.9%
Last updated: archived
Eligibility: Fallback benchmark identity is visible for context but excluded from default ranking.

Backfilled from GPT-5 via approved benchmark identity mapping map-gpt-5-4-nano-to-gpt-5.

92.9% percentile inside its fair comparison set

75.4%Raw benchmark value

Safety1 benchmarkn/a

MASK

SL · Safety · Rubric

Whether a model stays honest instead of covertly optimizing against the user.

Rank #20 · Source label: gpt-5-pro-2025-10-06

backfilledproxy backfilledBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Scale Labs
Raw value: 86%
Percentile: 74.6%
Last updated: recent
Eligibility: Fallback benchmark identity is visible for context but excluded from default ranking.

74.6% percentile inside its fair comparison set

86%Raw benchmark value

Embeddings / retrieval1 benchmarkn/a

Retrieval

MTEB · Embeddings / retrieval · Retrieval

It is one of the few direct signals for retrieval stacks, where embedding quality matters more than chat style.

Rank #4 · Source label: gpt-5

backfilledproxy backfilledBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: MTEB
Raw value: 58.8 ndcg
Percentile: 100%
Last updated: archived
Eligibility: Fallback benchmark identity is visible for context but excluded from default ranking.

Embedding endpoint score. Backfilled from GPT-5 via approved benchmark identity mapping map-gpt-5-4-nano-to-gpt-5.

100% percentile inside its fair comparison set

58.8 ndcgRaw benchmark value

Multilingual20 benchmarksn/a

Text Arena · Chinese

AR · Multilingual · Human

Observed user preference in Arena's Text Arena chinese leaderboard.

Rank #150 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,422
Percentile: 57.7%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: chinese. Source rank: #167. Votes: 3826. Organization: openai. License: Proprietary.

57.7% percentile inside its fair comparison set

1,422Raw benchmark valueCI 1,412 - 1,433

Text Arena · French

AR · Multilingual · Human

Observed user preference in Arena's Text Arena french leaderboard.

Rank #109 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,429
Percentile: 58%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: french. Source rank: #122. Votes: 2098. Organization: openai. License: Proprietary.

58% percentile inside its fair comparison set

1,429Raw benchmark valueCI 1,414 - 1,445

Text Arena · German

AR · Multilingual · Human

Observed user preference in Arena's Text Arena german leaderboard.

Rank #109 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,394
Percentile: 60.3%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: german. Source rank: #125. Votes: 931. Organization: openai. License: Proprietary.

60.3% percentile inside its fair comparison set

1,394Raw benchmark valueCI 1,373 - 1,415

Text Arena · Japanese

AR · Multilingual · Human

Observed user preference in Arena's Text Arena japanese leaderboard.

Rank #92 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,373
Percentile: 62.6%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: japanese. Source rank: #107. Votes: 603. Organization: openai. License: Proprietary.

62.6% percentile inside its fair comparison set

1,373Raw benchmark valueCI 1,346 - 1,399

Text Arena · Korean

AR · Multilingual · Human

Observed user preference in Arena's Text Arena korean leaderboard.

Rank #124 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,344
Percentile: 49.6%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: korean. Source rank: #141. Votes: 1031. Organization: openai. License: Proprietary.

49.6% percentile inside its fair comparison set

1,344Raw benchmark valueCI 1,324 - 1,364

Text Arena · Non English

AR · Multilingual · Human

Observed user preference in Arena's Text Arena non_english leaderboard.

Rank #141 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,386
Percentile: 62.8%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: non_english. Source rank: #158. Votes: 33441. Organization: openai. License: Proprietary.

62.8% percentile inside its fair comparison set

1,386Raw benchmark valueCI 1,382 - 1,391

Text Arena · Polish

AR · Multilingual · Human

Observed user preference in Arena's Text Arena polish leaderboard.

Rank #107 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,402
Percentile: 46.7%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: polish. Source rank: #122. Votes: 1130. Organization: openai. License: Proprietary.

46.7% percentile inside its fair comparison set

1,402Raw benchmark valueCI 1,383 - 1,420

Text Arena · Russian

AR · Multilingual · Human

Observed user preference in Arena's Text Arena russian leaderboard.

Rank #143 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,389
Percentile: 58.4%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: russian. Source rank: #161. Votes: 6875. Organization: openai. License: Proprietary.

58.4% percentile inside its fair comparison set

1,389Raw benchmark valueCI 1,381 - 1,397

Text Arena · Spanish

AR · Multilingual · Human

Observed user preference in Arena's Text Arena spanish leaderboard.

Rank #136 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,393
Percentile: 47.9%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: spanish. Source rank: #154. Votes: 1912. Organization: openai. License: Proprietary.

47.9% percentile inside its fair comparison set

1,393Raw benchmark valueCI 1,378 - 1,408

Text Arena · Chinese · No Style Control

AR · Multilingual · Human

Observed user preference in Arena's Text Arena chinese leaderboard.

Rank #164 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,391
Percentile: 53.7%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: chinese. Source rank: #182. Votes: 3826. Organization: openai. License: Proprietary.

53.7% percentile inside its fair comparison set

1,391Raw benchmark valueCI 1,381 - 1,402

Text Arena · French · No Style Control

AR · Multilingual · Human

Observed user preference in Arena's Text Arena french leaderboard.

Rank #132 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,396
Percentile: 49%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: french. Source rank: #147. Votes: 2098. Organization: openai. License: Proprietary.

49% percentile inside its fair comparison set

1,396Raw benchmark valueCI 1,381 - 1,411

Text Arena · German · No Style Control

AR · Multilingual · Human

Observed user preference in Arena's Text Arena german leaderboard.

Rank #131 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,368
Percentile: 52.2%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: german. Source rank: #149. Votes: 931. Organization: openai. License: Proprietary.

52.2% percentile inside its fair comparison set

1,368Raw benchmark valueCI 1,347 - 1,389

Text Arena · Japanese · No Style Control

AR · Multilingual · Human

Observed user preference in Arena's Text Arena japanese leaderboard.

Rank #110 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,341
Percentile: 55.1%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: japanese. Source rank: #126. Votes: 603. Organization: openai. License: Proprietary.

55.1% percentile inside its fair comparison set

1,341Raw benchmark valueCI 1,315 - 1,367

Text Arena · Korean · No Style Control

AR · Multilingual · Human

Observed user preference in Arena's Text Arena korean leaderboard.

Rank #133 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,322
Percentile: 45.9%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: korean. Source rank: #150. Votes: 1031. Organization: openai. License: Proprietary.

45.9% percentile inside its fair comparison set

1,322Raw benchmark valueCI 1,302 - 1,341

Text Arena · Non English · No Style Control

AR · Multilingual · Human

Observed user preference in Arena's Text Arena non_english leaderboard.

Rank #158 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,359
Percentile: 58.2%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: non_english. Source rank: #177. Votes: 33441. Organization: openai. License: Proprietary.

58.2% percentile inside its fair comparison set

1,359Raw benchmark valueCI 1,355 - 1,364

Text Arena · Polish · No Style Control

AR · Multilingual · Human

Observed user preference in Arena's Text Arena polish leaderboard.

Rank #131 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,365
Percentile: 34.7%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: polish. Source rank: #149. Votes: 1130. Organization: openai. License: Proprietary.

34.7% percentile inside its fair comparison set

1,365Raw benchmark valueCI 1,347 - 1,384

Text Arena · Russian · No Style Control

AR · Multilingual · Human

Observed user preference in Arena's Text Arena russian leaderboard.

Rank #157 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,362
Percentile: 54.3%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: russian. Source rank: #176. Votes: 6875. Organization: openai. License: Proprietary.

54.3% percentile inside its fair comparison set

1,362Raw benchmark valueCI 1,354 - 1,370

Text Arena · Spanish · No Style Control

AR · Multilingual · Human

Observed user preference in Arena's Text Arena spanish leaderboard.

Rank #151 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,370
Percentile: 42.1%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: spanish. Source rank: #168. Votes: 1912. Organization: openai. License: Proprietary.

42.1% percentile inside its fair comparison set

1,370Raw benchmark valueCI 1,355 - 1,385

Vision Arena · Chinese

AR · Multilingual · Human

Observed user preference in Arena's Vision Arena chinese leaderboard.

Rank #61 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,236
Percentile: 42.3%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: chinese. Source rank: #70. Votes: 1734. Organization: openai. License: Proprietary.

42.3% percentile inside its fair comparison set

1,236Raw benchmark valueCI 1,219 - 1,254

Vision Arena · Chinese · No Style Control

AR · Multilingual · Human

Observed user preference in Arena's Vision Arena chinese leaderboard.

Rank #66 · Source label: gpt-5.4-nano-high

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,232
Percentile: 37.5%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gpt-5.4-nano-high`. Category: chinese. Source rank: #76. Votes: 1734. Organization: openai. License: Proprietary.

37.5% percentile inside its fair comparison set

1,232Raw benchmark valueCI 1,215 - 1,249

Source links and registry checks

official

Terminal-Bench

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Artificial Analysis

Oct 3, 2026

source →

official

LiveBench

Oct 3, 2026

source →

GPT-5.4 nano

Current snapshot.

Source-linked scores by benchmark

Source links and registry checks

Loading model evidence.

GPT-5.4 nano

Current snapshot.

Source-linked scores by benchmark

Source links and registry checks