Model profile · Google

Gemini 2.5 Pro

Closed weightsfrontier · registry tag 2026 flagship

Thin verified coverage

Reads as thin verified coverage across the resolved source data.

Visible coverage: 36.3%
Verified coverage: 36.3%
Spread: 100%
Last verified: Jun 20, 2026

48%bench fit

textcodevisiondocumentaudiosearch11 aliases48 official source links

Open compare

Data version

Current snapshot.

Data version Jun 20, 2026Model list checked9 providers · 1081 tracked modelsPage refreshed Jul 5, 2026

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

Source-linked scores by benchmark

Each row keeps the benchmark source, source type, raw metric, and percentile inside its fair comparison set.

Thin verified coverageThis model currently reads as thin verified coverage across the resolved source data.

Chat / text29 benchmarks71.8%

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 #102 · Source label: Gemini 2.5 Pro Preview (May' 25)

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 22
Percentile: 74.4%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `intelligenceIndex`.

74.4% percentile inside its fair comparison set

22Raw 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 #16 · Source label: Gemini 2.5 Pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 39%
Percentile: 95%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `omniscienceAccuracy`.

95% percentile inside its fair comparison set

39%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 #182 · Source label: Gemini 2.5 Pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 12.6%
Percentile: 39.3%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `omniscienceNonHallucination`.

39.3% percentile inside its fair comparison set

12.6%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 #99 · Source label: Gemini 2.5 Pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 48.7%
Percentile: 68.9%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `ifbench`.

68.9% percentile inside its fair comparison set

48.7%Raw benchmark value

Blended 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 #228 · Source label: Gemini 2.5 Pro Preview (May' 25)

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: $3.4 /1M tokens
Percentile: 18.1%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `price1mBlended0To3To1`.

18.1% percentile inside its fair comparison set

$3.4 /1M tokensRaw 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 #210 · Source label: Gemini 2.5 Pro Preview (May' 25)

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: $1.3 /1M input tokens
Percentile: 24.6%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `price1mInputTokens`.

24.6% percentile inside its fair comparison set

$1.3 /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 #237 · Source label: Gemini 2.5 Pro Preview (May' 25)

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: $10 /1M output tokens
Percentile: 15.2%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `price1mOutputTokens`.

15.2% percentile inside its fair comparison set

$10 /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 #66 · Source label: Gemini 2.5 Pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 133.9 tokens/s
Percentile: 69%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

69% percentile inside its fair comparison set

133.9 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 #188 · Source label: Gemini 2.5 Pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 19.34s
Percentile: 11%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `medianTimeToFirstTokenSeconds`.

11% percentile inside its fair comparison set

19.34sRaw 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 #126 · Source label: Gemini 2.5 Pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 19.34s
Percentile: 40.5%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `medianTimeToFirstAnswerTokenSeconds`.

40.5% percentile inside its fair comparison set

19.34sRaw benchmark value

Openness Index

AA · Chat / text · Combined

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

Rank #178 · Source label: Gemini 2.5 Pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 6
Percentile: 7.5%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `opennessBreakdown.opennessIndex`.

7.5% percentile inside its fair comparison set

6Raw 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 #41 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,446
Percentile: 87.7%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: overall. Source rank: #53. Votes: 124588. Organization: google. License: Proprietary.

87.7% percentile inside its fair comparison set

1,446Raw benchmark valueCI 1,443 - 1,448

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 #23 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,444
Percentile: 93.2%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: creative_writing. Source rank: #32. Votes: 17441. Organization: google. License: Proprietary.

93.2% percentile inside its fair comparison set

1,444Raw benchmark valueCI 1,439 - 1,449

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 #49 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,448
Percentile: 85.2%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: english. Source rank: #61. Votes: 59047. Organization: google. License: Proprietary.

85.2% percentile inside its fair comparison set

1,448Raw benchmark valueCI 1,445 - 1,451

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 #42 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,445
Percentile: 87.4%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: exclude_ties. Source rank: #54. Votes: 90993. Organization: google. License: Proprietary.

87.4% percentile inside its fair comparison set

1,445Raw benchmark valueCI 1,442 - 1,449

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 #50 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,458
Percentile: 84.9%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: hard_prompts. Source rank: #64. Votes: 64682. Organization: google. License: Proprietary.

84.9% percentile inside its fair comparison set

1,458Raw benchmark valueCI 1,455 - 1,461

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 #59 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,457
Percentile: 82.1%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: hard_prompts_english. Source rank: #75. Votes: 32491. Organization: google. License: Proprietary.

82.1% percentile inside its fair comparison set

1,457Raw benchmark valueCI 1,453 - 1,461

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 #36 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,440
Percentile: 89.2%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: instruction_following. Source rank: #47. Votes: 34270. Organization: google. License: Proprietary.

89.2% percentile inside its fair comparison set

1,440Raw benchmark valueCI 1,436 - 1,443

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 #35 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,457
Percentile: 88.8%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: longer_query. Source rank: #46. Votes: 32626. Organization: google. License: Proprietary.

88.8% percentile inside its fair comparison set

1,457Raw benchmark valueCI 1,453 - 1,461

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 #47 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,448
Percentile: 85.8%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: multi_turn. Source rank: #61. Votes: 21191. Organization: google. License: Proprietary.

85.8% percentile inside its fair comparison set

1,448Raw benchmark valueCI 1,443 - 1,453

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 #17 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,457
Percentile: 95.1%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: overall. Source rank: #21. Votes: 124588. Organization: google. License: Proprietary.

95.1% percentile inside its fair comparison set

1,457Raw benchmark valueCI 1,455 - 1,460

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 #13 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,454
Percentile: 96.3%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: creative_writing. Source rank: #15. Votes: 17441. Organization: google. License: Proprietary.

96.3% percentile inside its fair comparison set

1,454Raw benchmark valueCI 1,449 - 1,459

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 #22 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,457
Percentile: 93.5%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: english. Source rank: #27. Votes: 59047. Organization: google. License: Proprietary.

93.5% percentile inside its fair comparison set

1,457Raw benchmark valueCI 1,454 - 1,460

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 #17 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,460
Percentile: 95.1%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: exclude_ties. Source rank: #21. Votes: 90993. Organization: google. License: Proprietary.

95.1% percentile inside its fair comparison set

1,460Raw benchmark valueCI 1,456 - 1,463

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 #27 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,455
Percentile: 92%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: hard_prompts. Source rank: #34. Votes: 64682. Organization: google. License: Proprietary.

92% percentile inside its fair comparison set

1,455Raw benchmark valueCI 1,452 - 1,458

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 #40 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,451
Percentile: 88%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: hard_prompts_english. Source rank: #47. Votes: 32491. Organization: google. License: Proprietary.

88% percentile inside its fair comparison set

1,451Raw benchmark valueCI 1,447 - 1,455

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 #27 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,437
Percentile: 92%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: instruction_following. Source rank: #34. Votes: 34270. Organization: google. License: Proprietary.

92% percentile inside its fair comparison set

1,437Raw benchmark valueCI 1,433 - 1,441

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 #26 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,449
Percentile: 91.8%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: longer_query. Source rank: #33. Votes: 32626. Organization: google. License: Proprietary.

91.8% percentile inside its fair comparison set

1,449Raw benchmark valueCI 1,445 - 1,453

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 #29 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,452
Percentile: 91.3%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: multi_turn. Source rank: #37. Votes: 21191. Organization: google. License: Proprietary.

91.3% percentile inside its fair comparison set

1,452Raw benchmark valueCI 1,447 - 1,457

Coding14 benchmarks36.8%

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 #77 · Source label: Gemini 2.5 Pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 26.5%
Percentile: 74.8%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `terminalbenchHard`.

74.8% percentile inside its fair comparison set

26.5%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 #43 · Source label: Gemini 2.5 Pro Preview (May' 25)

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 41.6%
Percentile: 88.6%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `scicode`.

88.6% percentile inside its fair comparison set

41.6%Raw benchmark value

Code 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 #68 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,204
Percentile: 8.2%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: overall. Source rank: #84. Votes: 3300. Organization: google. License: Proprietary.

8.2% percentile inside its fair comparison set

1,204Raw benchmark valueCI 1,191 - 1,217

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 #68 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,204
Percentile: 8.2%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: webdev. Source rank: #84. Votes: 3300. Organization: google. License: Proprietary.

8.2% percentile inside its fair comparison set

1,204Raw benchmark valueCI 1,191 - 1,217

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 #70 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,211
Percentile: 5.5%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: webdev-html. Source rank: #85. Votes: 3300. Organization: google. License: Proprietary.

5.5% percentile inside its fair comparison set

1,211Raw benchmark valueCI 1,197 - 1,225

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 #47 · Source label: google/gemini-2.5-pro-preview-03-25

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 79.2%
Percentile: 48.9%
Last updated: recent
Eligibility: headline eligible

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

48.9% percentile inside its fair comparison set

79.2%Raw benchmark valueCI 77% - 81.3%

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 #50 · Source label: google/gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 54.4%
Percentile: 9.3%
Last updated: recent
Eligibility: headline eligible

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

9.3% percentile inside its fair comparison set

54.4%Raw benchmark valueCI 50% - 58.8%

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 #47 · Source label: google/gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 0.4%
Percentile: 6.1%
Last updated: recent
Eligibility: headline eligible

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

6.1% percentile inside its fair comparison set

0.4%Raw benchmark valueCI 0% - 1.2%

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 #71 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,465
Percentile: 78.1%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: coding. Source rank: #89. Votes: 26567. Organization: google. License: Proprietary.

78.1% percentile inside its fair comparison set

1,465Raw benchmark valueCI 1,461 - 1,470

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 #47 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,452
Percentile: 85.6%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: coding. Source rank: #57. Votes: 26567. Organization: google. License: Proprietary.

85.6% percentile inside its fair comparison set

1,452Raw benchmark valueCI 1,448 - 1,457

IOI

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 #18 · Source label: google/gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 17.1%
Percentile: 61.4%
Last updated: recent
Eligibility: headline eligible

Parsed from Vals AI BenchmarkView overall scores. Vals slug: ioi; provider: Google.

61.4% percentile inside its fair comparison set

17.1%Raw benchmark valueCI 5% - 29.2%

Code Arena · Image To Webdev

AR · Coding · Human

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

Rank #18 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,276
Percentile: 0%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: image_to_webdev. Source rank: #23. Votes: 1185. Organization: google. License: Proprietary.

0% percentile inside its fair comparison set

1,276Raw benchmark valueCI 1,257 - 1,295

Terminal-Bench 2.0

TERMINAL-BENCH · Coding · Objective

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

Rank #19 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Terminal-Bench
Raw value: 32.6%
Percentile: 40%
Last updated: archived
Eligibility: headline eligible

Parsed from the public Terminal-Bench 2.0 verified leaderboard. Collapse policy: highest verified score per canonical model. Selected agent: Terminus 2 (2.0.0). Display model: Gemini 2.5 Pro. Integration method: API. Agent URL: https://www.tbench.ai/terminus. Reported stderr: 1.542 percentage points.

40% percentile inside its fair comparison set

32.6%Raw benchmark value

SWE-Bench Verified

OFF · Coding · Objective

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

Rank #6

Official company resultmanual verifiedmanual verified

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Provider official evals
Raw value: 63.8%
Percentile: 0%
Last updated: archived
Eligibility: headline eligible

Verified against Google's March 25, 2025 Gemini thinking models update post. Uses the published SWE-Bench Verified figure for Gemini 2.5 Pro with Google's stated custom agent setup. Checked 2026-04-29. Verification: manual_public_page_verification.

0% percentile inside its fair comparison set

63.8%Raw benchmark value

Reasoning / math / science11 benchmarks66.7%

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 #55 · Source label: Gemini 2.5 Pro Preview (May' 25)

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 15.4%
Percentile: 85.4%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `hle`.

85.4% percentile inside its fair comparison set

15.4%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 #52 · Source label: Gemini 2.5 Pro Preview (May' 25)

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 82.2%
Percentile: 86.4%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `gpqa`.

86.4% percentile inside its fair comparison set

82.2%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 #30 · Source label: Gemini 2.5 Pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 2.6%
Percentile: 90.7%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `critpt`.

90.7% percentile inside its fair comparison set

2.6%Raw benchmark value

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 #44 · Source label: google/gemini-2.5-pro-exp-03-25

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 80.8%
Percentile: 51.7%
Last updated: recent
Eligibility: headline eligible

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

51.7% percentile inside its fair comparison set

80.8%Raw benchmark valueCI 76.9% - 84.7%

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 #42 · Source label: google/gemini-2.5-pro-exp-03-25

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 84.1%
Percentile: 53.9%
Last updated: recent
Eligibility: headline eligible

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

53.9% percentile inside its fair comparison set

84.1%Raw benchmark valueCI 83.4% - 84.8%

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 #41 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,442
Percentile: 87.3%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: math. Source rank: #52. Votes: 7644. Organization: google. License: Proprietary.

87.3% percentile inside its fair comparison set

1,442Raw benchmark valueCI 1,435 - 1,449

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 #27 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,450
Percentile: 91.7%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: math. Source rank: #34. Votes: 7644. Organization: google. License: Proprietary.

91.7% percentile inside its fair comparison set

1,450Raw benchmark valueCI 1,443 - 1,457

TutorBench

SL · Reasoning / math / science · Rubric

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

Rank #2

verified runtimeexact direct

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Scale Labs
Raw value: 55.7%
Percentile: 90%
Last updated: recent
Eligibility: headline eligible

90% percentile inside its fair comparison set

55.7%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 #10

verified runtimeexact direct

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Scale Labs
Raw value: 45.1%
Percentile: 10%
Last updated: recent
Eligibility: headline eligible

10% percentile inside its fair comparison set

45.1%Raw benchmark value

EnigmaEval

SL · Reasoning / math / science · Rubric

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

Rank #3

verified runtimeexact direct

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Scale Labs
Raw value: 66%
Percentile: 86.7%
Last updated: aging
Eligibility: headline eligible

86.7% percentile inside its fair comparison set

66%Raw benchmark value

Humanity's Last Exam

OFF · Reasoning / math / science · Objective

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

Rank #8

Official company resultmanual verifiedmanual verified

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Provider official evals
Raw value: 21.6%
Percentile: 0%
Last updated: archived
Eligibility: headline eligible

Verified against the public Gemini 2.5 Pro model card PDF dated June 27, 2025. Uses the published no-tools Humanity's Last Exam figure. Checked 2026-04-29. Verification: manual_public_page_verification.

0% percentile inside its fair comparison set

21.6%Raw benchmark value

Professional reasoning25 benchmarks79%

LegalBench

VALS-AI · Professional reasoning · Objective

Academic legal reasoning tasks.

Rank #18 · Source label: google/gemini-2.5-pro-exp-03-25

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 84.3%
Percentile: 81.1%
Last updated: recent
Eligibility: headline eligible

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

81.1% percentile inside its fair comparison set

84.3%Raw benchmark valueCI 83.5% - 85.1%

TaxEval v2

VALS-AI · Professional reasoning · Objective

Answer quality on tax questions and responses.

Rank #38 · Source label: google/gemini-2.5-pro-exp-03-25

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 72.9%
Percentile: 59.3%
Last updated: recent
Eligibility: headline eligible

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

59.3% percentile inside its fair comparison set

72.9%Raw benchmark valueCI 71.2% - 74.6%

MedCode

VALS-AI · Professional reasoning · Objective

Medical billing support and coding tasks.

Rank #10 · Source label: google/gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 50.6%
Percentile: 82.4%
Last updated: recent
Eligibility: headline eligible

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

82.4% percentile inside its fair comparison set

50.6%Raw benchmark valueCI 46.4% - 54.7%

MedScribe

VALS-AI · Professional reasoning · Objective

Administrative documentation support for doctors.

Rank #37 · Source label: google/gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 73.6%
Percentile: 28%
Last updated: recent
Eligibility: headline eligible

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

28% percentile inside its fair comparison set

73.6%Raw benchmark valueCI 69.8% - 77.3%

Text Arena · Expert

AR · Professional reasoning · Human

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

Rank #50 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,461
Percentile: 82.2%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: expert. Source rank: #64. Votes: 7689. Organization: google. License: Proprietary.

82.2% percentile inside its fair comparison set

1,461Raw benchmark valueCI 1,454 - 1,469

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 #51 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,436
Percentile: 84.3%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: industry_business_and_management_and_financial_operations. Source rank: #67. Votes: 22479. Organization: google. License: Proprietary.

84.3% percentile inside its fair comparison set

1,436Raw benchmark valueCI 1,431 - 1,440

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 #30 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,430
Percentile: 91%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: industry_entertainment_and_sports_and_media. Source rank: #41. Votes: 22939. Organization: google. License: Proprietary.

91% percentile inside its fair comparison set

1,430Raw benchmark valueCI 1,426 - 1,435

Text Arena · Industry Legal And Government

AR · Professional reasoning · Human

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

Rank #25 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,468
Percentile: 91.9%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: industry_legal_and_government. Source rank: #34. Votes: 8797. Organization: google. License: Proprietary.

91.9% percentile inside its fair comparison set

1,468Raw benchmark valueCI 1,461 - 1,475

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 #37 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,470
Percentile: 88.9%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: industry_life_and_physical_and_social_science. Source rank: #47. Votes: 20439. Organization: google. License: Proprietary.

88.9% percentile inside its fair comparison set

1,470Raw benchmark valueCI 1,465 - 1,475

Text Arena · Industry Mathematical

AR · Professional reasoning · Human

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

Rank #42 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,449
Percentile: 86.7%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: industry_mathematical. Source rank: #53. Votes: 6619. Organization: google. License: Proprietary.

86.7% percentile inside its fair comparison set

1,449Raw benchmark valueCI 1,441 - 1,457

Text Arena · Industry Medicine And Healthcare

AR · Professional reasoning · Human

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

Rank #44 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,467
Percentile: 85.4%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: industry_medicine_and_healthcare. Source rank: #55. Votes: 7809. Organization: google. License: Proprietary.

85.4% percentile inside its fair comparison set

1,467Raw benchmark valueCI 1,460 - 1,475

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 #63 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,460
Percentile: 80.9%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: industry_software_and_it_services. Source rank: #79. Votes: 43931. Organization: google. License: Proprietary.

80.9% percentile inside its fair comparison set

1,460Raw benchmark valueCI 1,456 - 1,463

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 #27 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,445
Percentile: 92%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: industry_writing_and_literature_and_language. Source rank: #37. Votes: 27763. Organization: google. License: Proprietary.

92% percentile inside its fair comparison set

1,445Raw benchmark valueCI 1,441 - 1,450

Text Arena · Expert · No Style Control

AR · Professional reasoning · Human

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

Rank #33 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,455
Percentile: 88.4%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: expert. Source rank: #41. Votes: 7689. Organization: google. License: Proprietary.

88.4% percentile inside its fair comparison set

1,455Raw benchmark valueCI 1,447 - 1,462

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 #25 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,439
Percentile: 92.5%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: industry_business_and_management_and_financial_operations. Source rank: #30. Votes: 22479. Organization: google. License: Proprietary.

92.5% percentile inside its fair comparison set

1,439Raw benchmark valueCI 1,435 - 1,444

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 #16 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,441
Percentile: 95.4%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: industry_entertainment_and_sports_and_media. Source rank: #20. Votes: 22939. Organization: google. License: Proprietary.

95.4% percentile inside its fair comparison set

1,441Raw benchmark valueCI 1,436 - 1,445

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 #12 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,479
Percentile: 96.3%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: industry_legal_and_government. Source rank: #14. Votes: 8797. Organization: google. License: Proprietary.

96.3% percentile inside its fair comparison set

1,479Raw benchmark valueCI 1,472 - 1,485

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 #13 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,479
Percentile: 96.3%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: industry_life_and_physical_and_social_science. Source rank: #17. Votes: 20439. Organization: google. License: Proprietary.

96.3% percentile inside its fair comparison set

1,479Raw benchmark valueCI 1,474 - 1,484

Text Arena · Industry Mathematical · No Style Control

AR · Professional reasoning · Human

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

Rank #29 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,454
Percentile: 90.9%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: industry_mathematical. Source rank: #36. Votes: 6619. Organization: google. License: Proprietary.

90.9% percentile inside its fair comparison set

1,454Raw benchmark valueCI 1,446 - 1,462

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 #10 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,475
Percentile: 96.9%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: industry_medicine_and_healthcare. Source rank: #12. Votes: 7809. Organization: google. License: Proprietary.

96.9% percentile inside its fair comparison set

1,475Raw benchmark valueCI 1,468 - 1,483

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 #37 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,459
Percentile: 88.9%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: industry_software_and_it_services. Source rank: #46. Votes: 43931. Organization: google. License: Proprietary.

88.9% percentile inside its fair comparison set

1,459Raw benchmark valueCI 1,456 - 1,463

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 #14 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,454
Percentile: 96%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: industry_writing_and_literature_and_language. Source rank: #18. Votes: 27763. Organization: google. License: Proprietary.

96% percentile inside its fair comparison set

1,454Raw benchmark valueCI 1,449 - 1,458

SAGE

VALS-AI · Professional reasoning · Objective

Student Assessment with Generative Evaluation.

Rank #27 · Source label: google/gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 41.9%
Percentile: 42.2%
Last updated: recent
Eligibility: headline eligible

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

42.2% percentile inside its fair comparison set

41.9%Raw benchmark valueCI 35.2% - 48.6%

PRBench Legal

SL · Professional reasoning · Rubric

Applied legal reasoning on professional-domain tasks.

Rank #9

verified runtimeexact direct

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Scale Labs
Raw value: 41.4%
Percentile: 33.3%
Last updated: recent
Eligibility: headline eligible

33.3% percentile inside its fair comparison set

41.4%Raw benchmark value

Poker Agent

VALS-AI · Professional reasoning · Objective

Agent profit in poker-style strategic play.

Rank #13 · Source label: google/gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 1,032.6 score
Percentile: 25%
Last updated: archived
Eligibility: headline eligible

Parsed from Vals AI BenchmarkView overall scores. Vals slug: poker_agent; provider: unknown.

25% percentile inside its fair comparison set

1,032.6 scoreRaw benchmark valueCI 1,032.6 score - 1,032.6 score

Search / tool use3 benchmarks36.2%

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 #119 · Source label: Gemini 2.5 Pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 54.1%
Percentile: 61.8%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `tau2`.

61.8% percentile inside its fair comparison set

54.1%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 #26 · Source label: gemini-2.5-pro-grounding

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,142
Percentile: 16.7%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro-grounding`. Category: overall. Source rank: #25. Votes: 83776. Organization: google. License: Proprietary.

16.7% percentile inside its fair comparison set

1,142Raw benchmark valueCI 1,138 - 1,147

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 #22 · Source label: gemini-2.5-pro-grounding

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,142
Percentile: 30%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro-grounding`. Category: overall. Source rank: #24. Votes: 83776. Organization: google. License: Proprietary.

30% percentile inside its fair comparison set

1,142Raw benchmark valueCI 1,138 - 1,147

Long context2 benchmarks70.2%

Long Context Reasoning

AA · Long context · Objective

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

Rank #24 · Source label: Gemini 2.5 Pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 66%
Percentile: 92.7%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `lcr`.

92.7% percentile inside its fair comparison set

66%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 #47 · Source label: google/gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 60.8%
Percentile: 47.7%
Last updated: recent
Eligibility: headline eligible

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

47.7% percentile inside its fair comparison set

60.8%Raw benchmark valueCI 58.9% - 62.7%

Vision understanding24 benchmarks74.3%

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 #15 · Source label: Gemini 2.5 Pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 74.9%
Percentile: 89.6%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

89.6% percentile inside its fair comparison set

74.9%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 #24 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,246
Percentile: 78.9%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: overall. Source rank: #31. Votes: 89547. Organization: google. License: Proprietary.

78.9% percentile inside its fair comparison set

1,246Raw benchmark valueCI 1,241 - 1,251

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 #4 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,250
Percentile: 88.5%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: captioning. Source rank: #4. Votes: 806. Organization: google. License: Proprietary.

88.5% percentile inside its fair comparison set

1,250Raw benchmark valueCI 1,227 - 1,273

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 #15 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,257
Percentile: 74.5%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: creative_writing_vision. Source rank: #19. Votes: 3512. Organization: google. License: Proprietary.

74.5% percentile inside its fair comparison set

1,257Raw benchmark valueCI 1,245 - 1,269

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 #25 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,265
Percentile: 65.7%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: diagram. Source rank: #33. Votes: 9573. Organization: google. License: Proprietary.

65.7% percentile inside its fair comparison set

1,265Raw benchmark valueCI 1,258 - 1,272

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 #25 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,244
Percentile: 78%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: english. Source rank: #31. Votes: 42713. Organization: google. License: Proprietary.

78% percentile inside its fair comparison set

1,244Raw benchmark valueCI 1,237 - 1,251

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 #8 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,255
Percentile: 78.1%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: entity_recognition. Source rank: #7. Votes: 878. Organization: google. License: Proprietary.

78.1% percentile inside its fair comparison set

1,255Raw benchmark valueCI 1,233 - 1,277

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 #25 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,272
Percentile: 64.7%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: homework. Source rank: #32. Votes: 5452. Organization: google. License: Proprietary.

64.7% percentile inside its fair comparison set

1,272Raw benchmark valueCI 1,264 - 1,281

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 #19 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,245
Percentile: 63.3%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: humor. Source rank: #24. Votes: 3094. Organization: google. License: Proprietary.

63.3% percentile inside its fair comparison set

1,245Raw benchmark valueCI 1,231 - 1,259

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 #24 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,258
Percentile: 67.1%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: ocr. Source rank: #31. Votes: 32588. Organization: google. License: Proprietary.

67.1% percentile inside its fair comparison set

1,258Raw benchmark valueCI 1,253 - 1,262

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 #21 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,262
Percentile: 81.7%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: overall. Source rank: #27. Votes: 89547. Organization: google. License: Proprietary.

81.7% percentile inside its fair comparison set

1,262Raw benchmark valueCI 1,257 - 1,267

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 #3 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,301
Percentile: 92.3%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: captioning. Source rank: #3. Votes: 806. Organization: google. License: Proprietary.

92.3% percentile inside its fair comparison set

1,301Raw benchmark valueCI 1,280 - 1,323

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 #16 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,270
Percentile: 72.7%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: creative_writing_vision. Source rank: #20. Votes: 3512. Organization: google. License: Proprietary.

72.7% percentile inside its fair comparison set

1,270Raw benchmark valueCI 1,259 - 1,282

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 #21 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,271
Percentile: 71.4%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: diagram. Source rank: #27. Votes: 9573. Organization: google. License: Proprietary.

71.4% percentile inside its fair comparison set

1,271Raw benchmark valueCI 1,264 - 1,278

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 #22 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,261
Percentile: 80.7%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: english. Source rank: #27. Votes: 42713. Organization: google. License: Proprietary.

80.7% percentile inside its fair comparison set

1,261Raw benchmark valueCI 1,255 - 1,268

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 #8 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,283
Percentile: 78.1%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: entity_recognition. Source rank: #9. Votes: 878. Organization: google. License: Proprietary.

78.1% percentile inside its fair comparison set

1,283Raw benchmark valueCI 1,262 - 1,305

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 #22 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,281
Percentile: 69.1%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: homework. Source rank: #28. Votes: 5452. Organization: google. License: Proprietary.

69.1% percentile inside its fair comparison set

1,281Raw benchmark valueCI 1,273 - 1,290

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 #20 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,260
Percentile: 61.2%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: humor. Source rank: #26. Votes: 3094. Organization: google. License: Proprietary.

61.2% percentile inside its fair comparison set

1,260Raw benchmark valueCI 1,246 - 1,274

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 #21 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,269
Percentile: 71.4%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: ocr. Source rank: #26. Votes: 32588. Organization: google. License: Proprietary.

71.4% percentile inside its fair comparison set

1,269Raw benchmark valueCI 1,265 - 1,274

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 #24 · Source label: google/gemini-2.5-pro-exp-03-25

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 81.3%
Percentile: 60.3%
Last updated: recent
Eligibility: headline eligible

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

60.3% percentile inside its fair comparison set

81.3%Raw benchmark valueCI 79.5% - 83.2%

VTB

SL · Vision understanding · Rubric

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

Rank #9

verified runtimeexact direct

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Scale Labs
Raw value: 11.8%
Percentile: 27.3%
Last updated: recent
Eligibility: headline eligible

27.3% percentile inside its fair comparison set

11.8%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 #1

verified runtimeexact direct

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Scale Labs
Raw value: 82%
Percentile: 100%
Last updated: aging
Eligibility: headline eligible

Manually verified from the official Scale Labs leaderboard.

100% percentile inside its fair comparison set

82%Raw benchmark value

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 #6 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,262
Percentile: 84.4%
Last updated: archived
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: creative_writing. Source rank: #7. Votes: 2418. Organization: google. License: Proprietary.

84.4% percentile inside its fair comparison set

1,262Raw benchmark valueCI 1,248 - 1,276

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 #6 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,277
Percentile: 84.4%
Last updated: archived
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: creative_writing. Source rank: #7. Votes: 2418. Organization: google. License: Proprietary.

84.4% percentile inside its fair comparison set

1,277Raw benchmark valueCI 1,264 - 1,291

Document understanding4 benchmarks59.4%

Document Arena

AR · Document understanding · Human

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

Rank #21 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,430
Percentile: 16.7%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: overall. Source rank: #24. Votes: 24980. Organization: google. License: Proprietary.

16.7% percentile inside its fair comparison set

1,430Raw benchmark valueCI 1,424 - 1,436

Document Arena · No Style Control

AR · Document understanding · Human

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

Rank #18 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,420
Percentile: 29.2%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: overall. Source rank: #21. Votes: 24980. Organization: google. License: Proprietary.

29.2% percentile inside its fair comparison set

1,420Raw benchmark valueCI 1,414 - 1,426

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 #6 · Source label: google/gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 68.9%
Percentile: 91.7%
Last updated: recent
Eligibility: headline eligible

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

91.7% percentile inside its fair comparison set

68.9%Raw benchmark valueCI 67.1% - 70.7%

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 #1

verified runtimeexact direct

Raw row drilldownsource row, percentile, last updated, eligibility

Source: OpenCompass
Raw value: 78.3%
Percentile: 100%
Last updated: recent
Eligibility: headline eligible

Gemini leads document/multimodal coverage here.

100% percentile inside its fair comparison set

78.3%Raw benchmark value

Safety1 benchmark38.5%

MASK

SL · Safety · Rubric

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

Rank #9

verified runtimeexact direct

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Scale Labs
Raw value: 55.7%
Percentile: 38.5%
Last updated: recent
Eligibility: headline eligible

38.5% percentile inside its fair comparison set

55.7%Raw benchmark value

Embeddings / retrieval1 benchmark60%

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 #5

verified runtimeexact direct

Raw row drilldownsource row, percentile, last updated, eligibility

Source: MTEB
Raw value: 57.9 ndcg
Percentile: 60%
Last updated: aging
Eligibility: headline eligible

60% percentile inside its fair comparison set

57.9 ndcgRaw benchmark value

Audio dialogue / speech1 benchmark39%

Text to Speech

AA · Audio dialogue / speech · Human

Observed user preference over generated speech samples.

Rank #51 · Source label: Gemini 2.5 Pro (Dec 2025)

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 1,040
Percentile: 39%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public media leaderboard row `Gemini 2.5 Pro (Dec 2025)`.

39% percentile inside its fair comparison set

1,040Raw benchmark value

Multilingual16 benchmarks88.2%

Text Arena · Chinese

AR · Multilingual · Human

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

Rank #34 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,491
Percentile: 88.8%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: chinese. Source rank: #43. Votes: 6483. Organization: google. License: Proprietary.

88.8% percentile inside its fair comparison set

1,491Raw benchmark valueCI 1,483 - 1,499

Text Arena · French

AR · Multilingual · Human

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

Rank #42 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,464
Percentile: 81%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: french. Source rank: #53. Votes: 2202. Organization: google. License: Proprietary.

81% percentile inside its fair comparison set

1,464Raw benchmark valueCI 1,449 - 1,479

Text Arena · German

AR · Multilingual · Human

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

Rank #15 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,468
Percentile: 94.1%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: german. Source rank: #20. Votes: 2500. Organization: google. License: Proprietary.

94.1% percentile inside its fair comparison set

1,468Raw benchmark valueCI 1,455 - 1,482

Text Arena · Japanese

AR · Multilingual · Human

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

Rank #15 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,451
Percentile: 93.1%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: japanese. Source rank: #19. Votes: 1817. Organization: google. License: Proprietary.

93.1% percentile inside its fair comparison set

1,451Raw benchmark valueCI 1,436 - 1,467

Text Arena · Korean

AR · Multilingual · Human

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

Rank #29 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,415
Percentile: 86.5%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: korean. Source rank: #36. Votes: 2399. Organization: google. License: Proprietary.

86.5% percentile inside its fair comparison set

1,415Raw benchmark valueCI 1,401 - 1,428

Text Arena · Russian

AR · Multilingual · Human

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

Rank #33 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,453
Percentile: 88.9%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: russian. Source rank: #42. Votes: 10032. Organization: google. License: Proprietary.

88.9% percentile inside its fair comparison set

1,453Raw benchmark valueCI 1,446 - 1,459

Text Arena · Spanish

AR · Multilingual · Human

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

Rank #33 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,450
Percentile: 85%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: spanish. Source rank: #42. Votes: 2957. Organization: google. License: Proprietary.

85% percentile inside its fair comparison set

1,450Raw benchmark valueCI 1,437 - 1,462

Text Arena · Chinese · No Style Control

AR · Multilingual · Human

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

Rank #13 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,506
Percentile: 95.9%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: chinese. Source rank: #18. Votes: 6483. Organization: google. License: Proprietary.

95.9% percentile inside its fair comparison set

1,506Raw benchmark valueCI 1,498 - 1,515

Text Arena · French · No Style Control

AR · Multilingual · Human

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

Rank #19 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,472
Percentile: 91.7%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: french. Source rank: #23. Votes: 2202. Organization: google. License: Proprietary.

91.7% percentile inside its fair comparison set

1,472Raw benchmark valueCI 1,458 - 1,487

Text Arena · German · No Style Control

AR · Multilingual · Human

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

Rank #7 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,485
Percentile: 97.5%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: german. Source rank: #8. Votes: 2500. Organization: google. License: Proprietary.

97.5% percentile inside its fair comparison set

1,485Raw benchmark valueCI 1,472 - 1,499

Text Arena · Japanese · No Style Control

AR · Multilingual · Human

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

Rank #9 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,466
Percentile: 96.1%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: japanese. Source rank: #12. Votes: 1817. Organization: google. License: Proprietary.

96.1% percentile inside its fair comparison set

1,466Raw benchmark valueCI 1,450 - 1,482

Text Arena · Korean · No Style Control

AR · Multilingual · Human

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

Rank #12 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,434
Percentile: 94.7%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: korean. Source rank: #15. Votes: 2399. Organization: google. License: Proprietary.

94.7% percentile inside its fair comparison set

1,434Raw benchmark valueCI 1,421 - 1,448

Text Arena · Russian · No Style Control

AR · Multilingual · Human

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

Rank #19 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,460
Percentile: 93.8%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: russian. Source rank: #24. Votes: 10032. Organization: google. License: Proprietary.

93.8% percentile inside its fair comparison set

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

Text Arena · Spanish · No Style Control

AR · Multilingual · Human

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

Rank #9 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,468
Percentile: 96.3%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: spanish. Source rank: #11. Votes: 2957. Organization: google. License: Proprietary.

96.3% percentile inside its fair comparison set

1,468Raw benchmark valueCI 1,456 - 1,481

Vision Arena · Chinese

AR · Multilingual · Human

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

Rank #31 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,266
Percentile: 61%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: chinese. Source rank: #40. Votes: 3057. Organization: google. License: Proprietary.

61% percentile inside its fair comparison set

1,266Raw benchmark valueCI 1,249 - 1,283

Vision Arena · Chinese · No Style Control

AR · Multilingual · Human

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

Rank #27 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,290
Percentile: 66.2%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: chinese. Source rank: #34. Votes: 3057. Organization: google. License: Proprietary.

66.2% percentile inside its fair comparison set

1,290Raw benchmark valueCI 1,273 - 1,306

Source links and registry checks

official

Google Gemini models docs

Jun 20, 2026

source →

official

Arena

Jun 20, 2026

source →

official

Arena

Jun 20, 2026

source →

official

Arena

Jun 20, 2026

source →

official

Arena

Jun 20, 2026

source →

official

Arena

Jun 20, 2026

source →

official

Arena

Jun 20, 2026

source →

official

Arena

Jun 20, 2026

source →

official

Arena

Jun 20, 2026

source →

official

Arena

Jun 20, 2026

source →

official

Arena

Jun 20, 2026

source →

official

Arena

Jun 20, 2026

source →

official

Arena

Jun 20, 2026

source →

official

Arena

Jun 20, 2026

source →

official

Arena

Jun 20, 2026

source →

official

Arena

Jun 20, 2026

source →

official

Arena

Jun 20, 2026

source →

official

Arena

Jun 20, 2026

source →

official

Arena

Jun 20, 2026

source →

official

Arena

Jun 20, 2026

source →

official

Arena

Jun 20, 2026

source →

official

Arena

Jun 20, 2026

source →

official

Arena

Jun 20, 2026

source →

official

Arena

Jun 20, 2026

source →

official

Arena

Jun 20, 2026

source →

official

Arena

Jun 20, 2026

source →

official

Arena

Jun 20, 2026

source →

official

Arena

Jun 20, 2026

source →

official

Arena

Jun 20, 2026

source →

official

Arena

Jun 20, 2026

source →

official

Arena

Jun 20, 2026

source →

official

Arena

Jun 20, 2026

source →

official

Arena

Jun 20, 2026

source →

official

Arena

Jun 20, 2026

source →

official

Arena

Jun 20, 2026

source →

official

Arena

Jun 20, 2026

source →

official

Arena

Jun 20, 2026

source →

official

Arena

Jun 20, 2026

source →

official

Arena

Jun 20, 2026

source →

official

Arena

Jun 20, 2026

source →

official

Arena

Jun 20, 2026

source →

official

Arena

Jun 20, 2026

source →

official

Arena

Jun 20, 2026

source →

official

Arena

Jun 20, 2026

source →

official

Arena

Jun 20, 2026

source →

official

Artificial Analysis

Jun 20, 2026

source →

official

Artificial Analysis

Jun 20, 2026

source →

official

Terminal-Bench

Jun 20, 2026

source →

Model profile · Google

Gemini 2.5 Pro

Closed weightsfrontier · registry tag 2026 flagship

Thin verified coverage

Reads as thin verified coverage across the resolved source data.

Visible coverage: 36.3%
Verified coverage: 36.3%
Spread: 100%
Last verified: Jun 20, 2026

48%bench fit

textcodevisiondocumentaudiosearch11 aliases48 official source links

Open compare

Data version

Current snapshot.

Data version Jun 20, 2026Model list checked9 providers · 1081 tracked modelsPage refreshed Jul 5, 2026

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

Source-linked scores by benchmark

Each row keeps the benchmark source, source type, raw metric, and percentile inside its fair comparison set.

Thin verified coverageThis model currently reads as thin verified coverage across the resolved source data.

Chat / text29 benchmarks71.8%

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 #102 · Source label: Gemini 2.5 Pro Preview (May' 25)

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 22
Percentile: 74.4%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `intelligenceIndex`.

74.4% percentile inside its fair comparison set

22Raw 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 #16 · Source label: Gemini 2.5 Pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 39%
Percentile: 95%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `omniscienceAccuracy`.

95% percentile inside its fair comparison set

39%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 #182 · Source label: Gemini 2.5 Pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 12.6%
Percentile: 39.3%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `omniscienceNonHallucination`.

39.3% percentile inside its fair comparison set

12.6%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 #99 · Source label: Gemini 2.5 Pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 48.7%
Percentile: 68.9%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `ifbench`.

68.9% percentile inside its fair comparison set

48.7%Raw benchmark value

Blended 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 #228 · Source label: Gemini 2.5 Pro Preview (May' 25)

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: $3.4 /1M tokens
Percentile: 18.1%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `price1mBlended0To3To1`.

18.1% percentile inside its fair comparison set

$3.4 /1M tokensRaw 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 #210 · Source label: Gemini 2.5 Pro Preview (May' 25)

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: $1.3 /1M input tokens
Percentile: 24.6%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `price1mInputTokens`.

24.6% percentile inside its fair comparison set

$1.3 /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 #237 · Source label: Gemini 2.5 Pro Preview (May' 25)

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: $10 /1M output tokens
Percentile: 15.2%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `price1mOutputTokens`.

15.2% percentile inside its fair comparison set

$10 /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 #66 · Source label: Gemini 2.5 Pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 133.9 tokens/s
Percentile: 69%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

69% percentile inside its fair comparison set

133.9 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 #188 · Source label: Gemini 2.5 Pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 19.34s
Percentile: 11%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `medianTimeToFirstTokenSeconds`.

11% percentile inside its fair comparison set

19.34sRaw 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 #126 · Source label: Gemini 2.5 Pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 19.34s
Percentile: 40.5%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `medianTimeToFirstAnswerTokenSeconds`.

40.5% percentile inside its fair comparison set

19.34sRaw benchmark value

Openness Index

AA · Chat / text · Combined

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

Rank #178 · Source label: Gemini 2.5 Pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 6
Percentile: 7.5%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `opennessBreakdown.opennessIndex`.

7.5% percentile inside its fair comparison set

6Raw 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 #41 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,446
Percentile: 87.7%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: overall. Source rank: #53. Votes: 124588. Organization: google. License: Proprietary.

87.7% percentile inside its fair comparison set

1,446Raw benchmark valueCI 1,443 - 1,448

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 #23 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,444
Percentile: 93.2%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: creative_writing. Source rank: #32. Votes: 17441. Organization: google. License: Proprietary.

93.2% percentile inside its fair comparison set

1,444Raw benchmark valueCI 1,439 - 1,449

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 #49 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,448
Percentile: 85.2%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: english. Source rank: #61. Votes: 59047. Organization: google. License: Proprietary.

85.2% percentile inside its fair comparison set

1,448Raw benchmark valueCI 1,445 - 1,451

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 #42 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,445
Percentile: 87.4%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: exclude_ties. Source rank: #54. Votes: 90993. Organization: google. License: Proprietary.

87.4% percentile inside its fair comparison set

1,445Raw benchmark valueCI 1,442 - 1,449

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 #50 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,458
Percentile: 84.9%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: hard_prompts. Source rank: #64. Votes: 64682. Organization: google. License: Proprietary.

84.9% percentile inside its fair comparison set

1,458Raw benchmark valueCI 1,455 - 1,461

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 #59 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,457
Percentile: 82.1%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: hard_prompts_english. Source rank: #75. Votes: 32491. Organization: google. License: Proprietary.

82.1% percentile inside its fair comparison set

1,457Raw benchmark valueCI 1,453 - 1,461

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 #36 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,440
Percentile: 89.2%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: instruction_following. Source rank: #47. Votes: 34270. Organization: google. License: Proprietary.

89.2% percentile inside its fair comparison set

1,440Raw benchmark valueCI 1,436 - 1,443

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 #35 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,457
Percentile: 88.8%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: longer_query. Source rank: #46. Votes: 32626. Organization: google. License: Proprietary.

88.8% percentile inside its fair comparison set

1,457Raw benchmark valueCI 1,453 - 1,461

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 #47 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,448
Percentile: 85.8%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: multi_turn. Source rank: #61. Votes: 21191. Organization: google. License: Proprietary.

85.8% percentile inside its fair comparison set

1,448Raw benchmark valueCI 1,443 - 1,453

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 #17 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,457
Percentile: 95.1%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: overall. Source rank: #21. Votes: 124588. Organization: google. License: Proprietary.

95.1% percentile inside its fair comparison set

1,457Raw benchmark valueCI 1,455 - 1,460

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 #13 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,454
Percentile: 96.3%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: creative_writing. Source rank: #15. Votes: 17441. Organization: google. License: Proprietary.

96.3% percentile inside its fair comparison set

1,454Raw benchmark valueCI 1,449 - 1,459

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 #22 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,457
Percentile: 93.5%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: english. Source rank: #27. Votes: 59047. Organization: google. License: Proprietary.

93.5% percentile inside its fair comparison set

1,457Raw benchmark valueCI 1,454 - 1,460

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 #17 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,460
Percentile: 95.1%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: exclude_ties. Source rank: #21. Votes: 90993. Organization: google. License: Proprietary.

95.1% percentile inside its fair comparison set

1,460Raw benchmark valueCI 1,456 - 1,463

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 #27 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,455
Percentile: 92%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: hard_prompts. Source rank: #34. Votes: 64682. Organization: google. License: Proprietary.

92% percentile inside its fair comparison set

1,455Raw benchmark valueCI 1,452 - 1,458

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 #40 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,451
Percentile: 88%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: hard_prompts_english. Source rank: #47. Votes: 32491. Organization: google. License: Proprietary.

88% percentile inside its fair comparison set

1,451Raw benchmark valueCI 1,447 - 1,455

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 #27 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,437
Percentile: 92%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: instruction_following. Source rank: #34. Votes: 34270. Organization: google. License: Proprietary.

92% percentile inside its fair comparison set

1,437Raw benchmark valueCI 1,433 - 1,441

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 #26 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,449
Percentile: 91.8%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: longer_query. Source rank: #33. Votes: 32626. Organization: google. License: Proprietary.

91.8% percentile inside its fair comparison set

1,449Raw benchmark valueCI 1,445 - 1,453

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 #29 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,452
Percentile: 91.3%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: multi_turn. Source rank: #37. Votes: 21191. Organization: google. License: Proprietary.

91.3% percentile inside its fair comparison set

1,452Raw benchmark valueCI 1,447 - 1,457

Coding14 benchmarks36.8%

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 #77 · Source label: Gemini 2.5 Pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 26.5%
Percentile: 74.8%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `terminalbenchHard`.

74.8% percentile inside its fair comparison set

26.5%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 #43 · Source label: Gemini 2.5 Pro Preview (May' 25)

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 41.6%
Percentile: 88.6%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `scicode`.

88.6% percentile inside its fair comparison set

41.6%Raw benchmark value

Code 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 #68 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,204
Percentile: 8.2%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: overall. Source rank: #84. Votes: 3300. Organization: google. License: Proprietary.

8.2% percentile inside its fair comparison set

1,204Raw benchmark valueCI 1,191 - 1,217

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 #68 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,204
Percentile: 8.2%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: webdev. Source rank: #84. Votes: 3300. Organization: google. License: Proprietary.

8.2% percentile inside its fair comparison set

1,204Raw benchmark valueCI 1,191 - 1,217

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 #70 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,211
Percentile: 5.5%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: webdev-html. Source rank: #85. Votes: 3300. Organization: google. License: Proprietary.

5.5% percentile inside its fair comparison set

1,211Raw benchmark valueCI 1,197 - 1,225

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 #47 · Source label: google/gemini-2.5-pro-preview-03-25

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 79.2%
Percentile: 48.9%
Last updated: recent
Eligibility: headline eligible

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

48.9% percentile inside its fair comparison set

79.2%Raw benchmark valueCI 77% - 81.3%

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 #50 · Source label: google/gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 54.4%
Percentile: 9.3%
Last updated: recent
Eligibility: headline eligible

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

9.3% percentile inside its fair comparison set

54.4%Raw benchmark valueCI 50% - 58.8%

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 #47 · Source label: google/gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 0.4%
Percentile: 6.1%
Last updated: recent
Eligibility: headline eligible

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

6.1% percentile inside its fair comparison set

0.4%Raw benchmark valueCI 0% - 1.2%

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 #71 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,465
Percentile: 78.1%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: coding. Source rank: #89. Votes: 26567. Organization: google. License: Proprietary.

78.1% percentile inside its fair comparison set

1,465Raw benchmark valueCI 1,461 - 1,470

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 #47 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,452
Percentile: 85.6%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: coding. Source rank: #57. Votes: 26567. Organization: google. License: Proprietary.

85.6% percentile inside its fair comparison set

1,452Raw benchmark valueCI 1,448 - 1,457

IOI

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 #18 · Source label: google/gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 17.1%
Percentile: 61.4%
Last updated: recent
Eligibility: headline eligible

Parsed from Vals AI BenchmarkView overall scores. Vals slug: ioi; provider: Google.

61.4% percentile inside its fair comparison set

17.1%Raw benchmark valueCI 5% - 29.2%

Code Arena · Image To Webdev

AR · Coding · Human

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

Rank #18 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,276
Percentile: 0%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: image_to_webdev. Source rank: #23. Votes: 1185. Organization: google. License: Proprietary.

0% percentile inside its fair comparison set

1,276Raw benchmark valueCI 1,257 - 1,295

Terminal-Bench 2.0

TERMINAL-BENCH · Coding · Objective

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

Rank #19 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Terminal-Bench
Raw value: 32.6%
Percentile: 40%
Last updated: archived
Eligibility: headline eligible

40% percentile inside its fair comparison set

32.6%Raw benchmark value

SWE-Bench Verified

OFF · Coding · Objective

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

Rank #6

Official company resultmanual verifiedmanual verified

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Provider official evals
Raw value: 63.8%
Percentile: 0%
Last updated: archived
Eligibility: headline eligible

0% percentile inside its fair comparison set

63.8%Raw benchmark value

Reasoning / math / science11 benchmarks66.7%

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 #55 · Source label: Gemini 2.5 Pro Preview (May' 25)

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 15.4%
Percentile: 85.4%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `hle`.

85.4% percentile inside its fair comparison set

15.4%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 #52 · Source label: Gemini 2.5 Pro Preview (May' 25)

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 82.2%
Percentile: 86.4%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `gpqa`.

86.4% percentile inside its fair comparison set

82.2%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 #30 · Source label: Gemini 2.5 Pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 2.6%
Percentile: 90.7%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `critpt`.

90.7% percentile inside its fair comparison set

2.6%Raw benchmark value

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 #44 · Source label: google/gemini-2.5-pro-exp-03-25

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 80.8%
Percentile: 51.7%
Last updated: recent
Eligibility: headline eligible

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

51.7% percentile inside its fair comparison set

80.8%Raw benchmark valueCI 76.9% - 84.7%

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 #42 · Source label: google/gemini-2.5-pro-exp-03-25

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 84.1%
Percentile: 53.9%
Last updated: recent
Eligibility: headline eligible

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

53.9% percentile inside its fair comparison set

84.1%Raw benchmark valueCI 83.4% - 84.8%

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 #41 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,442
Percentile: 87.3%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: math. Source rank: #52. Votes: 7644. Organization: google. License: Proprietary.

87.3% percentile inside its fair comparison set

1,442Raw benchmark valueCI 1,435 - 1,449

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 #27 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,450
Percentile: 91.7%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: math. Source rank: #34. Votes: 7644. Organization: google. License: Proprietary.

91.7% percentile inside its fair comparison set

1,450Raw benchmark valueCI 1,443 - 1,457

TutorBench

SL · Reasoning / math / science · Rubric

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

Rank #2

verified runtimeexact direct

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Scale Labs
Raw value: 55.7%
Percentile: 90%
Last updated: recent
Eligibility: headline eligible

90% percentile inside its fair comparison set

55.7%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 #10

verified runtimeexact direct

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Scale Labs
Raw value: 45.1%
Percentile: 10%
Last updated: recent
Eligibility: headline eligible

10% percentile inside its fair comparison set

45.1%Raw benchmark value

EnigmaEval

SL · Reasoning / math / science · Rubric

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

Rank #3

verified runtimeexact direct

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Scale Labs
Raw value: 66%
Percentile: 86.7%
Last updated: aging
Eligibility: headline eligible

86.7% percentile inside its fair comparison set

66%Raw benchmark value

Humanity's Last Exam

OFF · Reasoning / math / science · Objective

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

Rank #8

Official company resultmanual verifiedmanual verified

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Provider official evals
Raw value: 21.6%
Percentile: 0%
Last updated: archived
Eligibility: headline eligible

0% percentile inside its fair comparison set

21.6%Raw benchmark value

Professional reasoning25 benchmarks79%

LegalBench

VALS-AI · Professional reasoning · Objective

Academic legal reasoning tasks.

Rank #18 · Source label: google/gemini-2.5-pro-exp-03-25

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 84.3%
Percentile: 81.1%
Last updated: recent
Eligibility: headline eligible

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

81.1% percentile inside its fair comparison set

84.3%Raw benchmark valueCI 83.5% - 85.1%

TaxEval v2

VALS-AI · Professional reasoning · Objective

Answer quality on tax questions and responses.

Rank #38 · Source label: google/gemini-2.5-pro-exp-03-25

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 72.9%
Percentile: 59.3%
Last updated: recent
Eligibility: headline eligible

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

59.3% percentile inside its fair comparison set

72.9%Raw benchmark valueCI 71.2% - 74.6%

MedCode

VALS-AI · Professional reasoning · Objective

Medical billing support and coding tasks.

Rank #10 · Source label: google/gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 50.6%
Percentile: 82.4%
Last updated: recent
Eligibility: headline eligible

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

82.4% percentile inside its fair comparison set

50.6%Raw benchmark valueCI 46.4% - 54.7%

MedScribe

VALS-AI · Professional reasoning · Objective

Administrative documentation support for doctors.

Rank #37 · Source label: google/gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 73.6%
Percentile: 28%
Last updated: recent
Eligibility: headline eligible

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

28% percentile inside its fair comparison set

73.6%Raw benchmark valueCI 69.8% - 77.3%

Text Arena · Expert

AR · Professional reasoning · Human

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

Rank #50 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,461
Percentile: 82.2%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: expert. Source rank: #64. Votes: 7689. Organization: google. License: Proprietary.

82.2% percentile inside its fair comparison set

1,461Raw benchmark valueCI 1,454 - 1,469

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 #51 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,436
Percentile: 84.3%
Last updated: recent
Eligibility: headline eligible

84.3% percentile inside its fair comparison set

1,436Raw benchmark valueCI 1,431 - 1,440

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 #30 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,430
Percentile: 91%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: industry_entertainment_and_sports_and_media. Source rank: #41. Votes: 22939. Organization: google. License: Proprietary.

91% percentile inside its fair comparison set

1,430Raw benchmark valueCI 1,426 - 1,435

Text Arena · Industry Legal And Government

AR · Professional reasoning · Human

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

Rank #25 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,468
Percentile: 91.9%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: industry_legal_and_government. Source rank: #34. Votes: 8797. Organization: google. License: Proprietary.

91.9% percentile inside its fair comparison set

1,468Raw benchmark valueCI 1,461 - 1,475

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 #37 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,470
Percentile: 88.9%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: industry_life_and_physical_and_social_science. Source rank: #47. Votes: 20439. Organization: google. License: Proprietary.

88.9% percentile inside its fair comparison set

1,470Raw benchmark valueCI 1,465 - 1,475

Text Arena · Industry Mathematical

AR · Professional reasoning · Human

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

Rank #42 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,449
Percentile: 86.7%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: industry_mathematical. Source rank: #53. Votes: 6619. Organization: google. License: Proprietary.

86.7% percentile inside its fair comparison set

1,449Raw benchmark valueCI 1,441 - 1,457

Text Arena · Industry Medicine And Healthcare

AR · Professional reasoning · Human

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

Rank #44 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,467
Percentile: 85.4%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: industry_medicine_and_healthcare. Source rank: #55. Votes: 7809. Organization: google. License: Proprietary.

85.4% percentile inside its fair comparison set

1,467Raw benchmark valueCI 1,460 - 1,475

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 #63 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,460
Percentile: 80.9%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: industry_software_and_it_services. Source rank: #79. Votes: 43931. Organization: google. License: Proprietary.

80.9% percentile inside its fair comparison set

1,460Raw benchmark valueCI 1,456 - 1,463

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 #27 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,445
Percentile: 92%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: industry_writing_and_literature_and_language. Source rank: #37. Votes: 27763. Organization: google. License: Proprietary.

92% percentile inside its fair comparison set

1,445Raw benchmark valueCI 1,441 - 1,450

Text Arena · Expert · No Style Control

AR · Professional reasoning · Human

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

Rank #33 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,455
Percentile: 88.4%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: expert. Source rank: #41. Votes: 7689. Organization: google. License: Proprietary.

88.4% percentile inside its fair comparison set

1,455Raw benchmark valueCI 1,447 - 1,462

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 #25 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,439
Percentile: 92.5%
Last updated: recent
Eligibility: headline eligible

92.5% percentile inside its fair comparison set

1,439Raw benchmark valueCI 1,435 - 1,444

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 #16 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,441
Percentile: 95.4%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: industry_entertainment_and_sports_and_media. Source rank: #20. Votes: 22939. Organization: google. License: Proprietary.

95.4% percentile inside its fair comparison set

1,441Raw benchmark valueCI 1,436 - 1,445

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 #12 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,479
Percentile: 96.3%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: industry_legal_and_government. Source rank: #14. Votes: 8797. Organization: google. License: Proprietary.

96.3% percentile inside its fair comparison set

1,479Raw benchmark valueCI 1,472 - 1,485

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 #13 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,479
Percentile: 96.3%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: industry_life_and_physical_and_social_science. Source rank: #17. Votes: 20439. Organization: google. License: Proprietary.

96.3% percentile inside its fair comparison set

1,479Raw benchmark valueCI 1,474 - 1,484

Text Arena · Industry Mathematical · No Style Control

AR · Professional reasoning · Human

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

Rank #29 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,454
Percentile: 90.9%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: industry_mathematical. Source rank: #36. Votes: 6619. Organization: google. License: Proprietary.

90.9% percentile inside its fair comparison set

1,454Raw benchmark valueCI 1,446 - 1,462

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 #10 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,475
Percentile: 96.9%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: industry_medicine_and_healthcare. Source rank: #12. Votes: 7809. Organization: google. License: Proprietary.

96.9% percentile inside its fair comparison set

1,475Raw benchmark valueCI 1,468 - 1,483

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 #37 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,459
Percentile: 88.9%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: industry_software_and_it_services. Source rank: #46. Votes: 43931. Organization: google. License: Proprietary.

88.9% percentile inside its fair comparison set

1,459Raw benchmark valueCI 1,456 - 1,463

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 #14 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,454
Percentile: 96%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: industry_writing_and_literature_and_language. Source rank: #18. Votes: 27763. Organization: google. License: Proprietary.

96% percentile inside its fair comparison set

1,454Raw benchmark valueCI 1,449 - 1,458

SAGE

VALS-AI · Professional reasoning · Objective

Student Assessment with Generative Evaluation.

Rank #27 · Source label: google/gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 41.9%
Percentile: 42.2%
Last updated: recent
Eligibility: headline eligible

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

42.2% percentile inside its fair comparison set

41.9%Raw benchmark valueCI 35.2% - 48.6%

PRBench Legal

SL · Professional reasoning · Rubric

Applied legal reasoning on professional-domain tasks.

Rank #9

verified runtimeexact direct

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Scale Labs
Raw value: 41.4%
Percentile: 33.3%
Last updated: recent
Eligibility: headline eligible

33.3% percentile inside its fair comparison set

41.4%Raw benchmark value

Poker Agent

VALS-AI · Professional reasoning · Objective

Agent profit in poker-style strategic play.

Rank #13 · Source label: google/gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 1,032.6 score
Percentile: 25%
Last updated: archived
Eligibility: headline eligible

Parsed from Vals AI BenchmarkView overall scores. Vals slug: poker_agent; provider: unknown.

25% percentile inside its fair comparison set

1,032.6 scoreRaw benchmark valueCI 1,032.6 score - 1,032.6 score

Search / tool use3 benchmarks36.2%

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 #119 · Source label: Gemini 2.5 Pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 54.1%
Percentile: 61.8%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `tau2`.

61.8% percentile inside its fair comparison set

54.1%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 #26 · Source label: gemini-2.5-pro-grounding

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,142
Percentile: 16.7%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro-grounding`. Category: overall. Source rank: #25. Votes: 83776. Organization: google. License: Proprietary.

16.7% percentile inside its fair comparison set

1,142Raw benchmark valueCI 1,138 - 1,147

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 #22 · Source label: gemini-2.5-pro-grounding

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,142
Percentile: 30%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro-grounding`. Category: overall. Source rank: #24. Votes: 83776. Organization: google. License: Proprietary.

30% percentile inside its fair comparison set

1,142Raw benchmark valueCI 1,138 - 1,147

Long context2 benchmarks70.2%

Long Context Reasoning

AA · Long context · Objective

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

Rank #24 · Source label: Gemini 2.5 Pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 66%
Percentile: 92.7%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `lcr`.

92.7% percentile inside its fair comparison set

66%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 #47 · Source label: google/gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 60.8%
Percentile: 47.7%
Last updated: recent
Eligibility: headline eligible

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

47.7% percentile inside its fair comparison set

60.8%Raw benchmark valueCI 58.9% - 62.7%

Vision understanding24 benchmarks74.3%

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 #15 · Source label: Gemini 2.5 Pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 74.9%
Percentile: 89.6%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

89.6% percentile inside its fair comparison set

74.9%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 #24 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,246
Percentile: 78.9%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: overall. Source rank: #31. Votes: 89547. Organization: google. License: Proprietary.

78.9% percentile inside its fair comparison set

1,246Raw benchmark valueCI 1,241 - 1,251

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 #4 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,250
Percentile: 88.5%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: captioning. Source rank: #4. Votes: 806. Organization: google. License: Proprietary.

88.5% percentile inside its fair comparison set

1,250Raw benchmark valueCI 1,227 - 1,273

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 #15 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,257
Percentile: 74.5%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: creative_writing_vision. Source rank: #19. Votes: 3512. Organization: google. License: Proprietary.

74.5% percentile inside its fair comparison set

1,257Raw benchmark valueCI 1,245 - 1,269

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 #25 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,265
Percentile: 65.7%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: diagram. Source rank: #33. Votes: 9573. Organization: google. License: Proprietary.

65.7% percentile inside its fair comparison set

1,265Raw benchmark valueCI 1,258 - 1,272

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 #25 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,244
Percentile: 78%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: english. Source rank: #31. Votes: 42713. Organization: google. License: Proprietary.

78% percentile inside its fair comparison set

1,244Raw benchmark valueCI 1,237 - 1,251

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 #8 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,255
Percentile: 78.1%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: entity_recognition. Source rank: #7. Votes: 878. Organization: google. License: Proprietary.

78.1% percentile inside its fair comparison set

1,255Raw benchmark valueCI 1,233 - 1,277

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 #25 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,272
Percentile: 64.7%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: homework. Source rank: #32. Votes: 5452. Organization: google. License: Proprietary.

64.7% percentile inside its fair comparison set

1,272Raw benchmark valueCI 1,264 - 1,281

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 #19 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,245
Percentile: 63.3%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: humor. Source rank: #24. Votes: 3094. Organization: google. License: Proprietary.

63.3% percentile inside its fair comparison set

1,245Raw benchmark valueCI 1,231 - 1,259

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 #24 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,258
Percentile: 67.1%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: ocr. Source rank: #31. Votes: 32588. Organization: google. License: Proprietary.

67.1% percentile inside its fair comparison set

1,258Raw benchmark valueCI 1,253 - 1,262

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 #21 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,262
Percentile: 81.7%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: overall. Source rank: #27. Votes: 89547. Organization: google. License: Proprietary.

81.7% percentile inside its fair comparison set

1,262Raw benchmark valueCI 1,257 - 1,267

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 #3 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,301
Percentile: 92.3%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: captioning. Source rank: #3. Votes: 806. Organization: google. License: Proprietary.

92.3% percentile inside its fair comparison set

1,301Raw benchmark valueCI 1,280 - 1,323

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 #16 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,270
Percentile: 72.7%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: creative_writing_vision. Source rank: #20. Votes: 3512. Organization: google. License: Proprietary.

72.7% percentile inside its fair comparison set

1,270Raw benchmark valueCI 1,259 - 1,282

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 #21 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,271
Percentile: 71.4%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: diagram. Source rank: #27. Votes: 9573. Organization: google. License: Proprietary.

71.4% percentile inside its fair comparison set

1,271Raw benchmark valueCI 1,264 - 1,278

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 #22 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,261
Percentile: 80.7%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: english. Source rank: #27. Votes: 42713. Organization: google. License: Proprietary.

80.7% percentile inside its fair comparison set

1,261Raw benchmark valueCI 1,255 - 1,268

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 #8 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,283
Percentile: 78.1%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: entity_recognition. Source rank: #9. Votes: 878. Organization: google. License: Proprietary.

78.1% percentile inside its fair comparison set

1,283Raw benchmark valueCI 1,262 - 1,305

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 #22 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,281
Percentile: 69.1%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: homework. Source rank: #28. Votes: 5452. Organization: google. License: Proprietary.

69.1% percentile inside its fair comparison set

1,281Raw benchmark valueCI 1,273 - 1,290

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 #20 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,260
Percentile: 61.2%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: humor. Source rank: #26. Votes: 3094. Organization: google. License: Proprietary.

61.2% percentile inside its fair comparison set

1,260Raw benchmark valueCI 1,246 - 1,274

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 #21 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,269
Percentile: 71.4%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: ocr. Source rank: #26. Votes: 32588. Organization: google. License: Proprietary.

71.4% percentile inside its fair comparison set

1,269Raw benchmark valueCI 1,265 - 1,274

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 #24 · Source label: google/gemini-2.5-pro-exp-03-25

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 81.3%
Percentile: 60.3%
Last updated: recent
Eligibility: headline eligible

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

60.3% percentile inside its fair comparison set

81.3%Raw benchmark valueCI 79.5% - 83.2%

VTB

SL · Vision understanding · Rubric

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

Rank #9

verified runtimeexact direct

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Scale Labs
Raw value: 11.8%
Percentile: 27.3%
Last updated: recent
Eligibility: headline eligible

27.3% percentile inside its fair comparison set

11.8%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 #1

verified runtimeexact direct

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Scale Labs
Raw value: 82%
Percentile: 100%
Last updated: aging
Eligibility: headline eligible

Manually verified from the official Scale Labs leaderboard.

100% percentile inside its fair comparison set

82%Raw benchmark value

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 #6 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,262
Percentile: 84.4%
Last updated: archived
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: creative_writing. Source rank: #7. Votes: 2418. Organization: google. License: Proprietary.

84.4% percentile inside its fair comparison set

1,262Raw benchmark valueCI 1,248 - 1,276

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 #6 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,277
Percentile: 84.4%
Last updated: archived
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: creative_writing. Source rank: #7. Votes: 2418. Organization: google. License: Proprietary.

84.4% percentile inside its fair comparison set

1,277Raw benchmark valueCI 1,264 - 1,291

Document understanding4 benchmarks59.4%

Document Arena

AR · Document understanding · Human

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

Rank #21 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,430
Percentile: 16.7%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: overall. Source rank: #24. Votes: 24980. Organization: google. License: Proprietary.

16.7% percentile inside its fair comparison set

1,430Raw benchmark valueCI 1,424 - 1,436

Document Arena · No Style Control

AR · Document understanding · Human

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

Rank #18 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,420
Percentile: 29.2%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: overall. Source rank: #21. Votes: 24980. Organization: google. License: Proprietary.

29.2% percentile inside its fair comparison set

1,420Raw benchmark valueCI 1,414 - 1,426

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 #6 · Source label: google/gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 68.9%
Percentile: 91.7%
Last updated: recent
Eligibility: headline eligible

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

91.7% percentile inside its fair comparison set

68.9%Raw benchmark valueCI 67.1% - 70.7%

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 #1

verified runtimeexact direct

Raw row drilldownsource row, percentile, last updated, eligibility

Source: OpenCompass
Raw value: 78.3%
Percentile: 100%
Last updated: recent
Eligibility: headline eligible

Gemini leads document/multimodal coverage here.

100% percentile inside its fair comparison set

78.3%Raw benchmark value

Safety1 benchmark38.5%

MASK

SL · Safety · Rubric

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

Rank #9

verified runtimeexact direct

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Scale Labs
Raw value: 55.7%
Percentile: 38.5%
Last updated: recent
Eligibility: headline eligible

38.5% percentile inside its fair comparison set

55.7%Raw benchmark value

Embeddings / retrieval1 benchmark60%

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 #5

verified runtimeexact direct

Raw row drilldownsource row, percentile, last updated, eligibility

Source: MTEB
Raw value: 57.9 ndcg
Percentile: 60%
Last updated: aging
Eligibility: headline eligible

60% percentile inside its fair comparison set

57.9 ndcgRaw benchmark value

Audio dialogue / speech1 benchmark39%

Text to Speech

AA · Audio dialogue / speech · Human

Observed user preference over generated speech samples.

Rank #51 · Source label: Gemini 2.5 Pro (Dec 2025)

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 1,040
Percentile: 39%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public media leaderboard row `Gemini 2.5 Pro (Dec 2025)`.

39% percentile inside its fair comparison set

1,040Raw benchmark value

Multilingual16 benchmarks88.2%

Text Arena · Chinese

AR · Multilingual · Human

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

Rank #34 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,491
Percentile: 88.8%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: chinese. Source rank: #43. Votes: 6483. Organization: google. License: Proprietary.

88.8% percentile inside its fair comparison set

1,491Raw benchmark valueCI 1,483 - 1,499

Text Arena · French

AR · Multilingual · Human

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

Rank #42 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,464
Percentile: 81%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: french. Source rank: #53. Votes: 2202. Organization: google. License: Proprietary.

81% percentile inside its fair comparison set

1,464Raw benchmark valueCI 1,449 - 1,479

Text Arena · German

AR · Multilingual · Human

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

Rank #15 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,468
Percentile: 94.1%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: german. Source rank: #20. Votes: 2500. Organization: google. License: Proprietary.

94.1% percentile inside its fair comparison set

1,468Raw benchmark valueCI 1,455 - 1,482

Text Arena · Japanese

AR · Multilingual · Human

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

Rank #15 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,451
Percentile: 93.1%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: japanese. Source rank: #19. Votes: 1817. Organization: google. License: Proprietary.

93.1% percentile inside its fair comparison set

1,451Raw benchmark valueCI 1,436 - 1,467

Text Arena · Korean

AR · Multilingual · Human

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

Rank #29 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,415
Percentile: 86.5%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: korean. Source rank: #36. Votes: 2399. Organization: google. License: Proprietary.

86.5% percentile inside its fair comparison set

1,415Raw benchmark valueCI 1,401 - 1,428

Text Arena · Russian

AR · Multilingual · Human

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

Rank #33 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,453
Percentile: 88.9%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: russian. Source rank: #42. Votes: 10032. Organization: google. License: Proprietary.

88.9% percentile inside its fair comparison set

1,453Raw benchmark valueCI 1,446 - 1,459

Text Arena · Spanish

AR · Multilingual · Human

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

Rank #33 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,450
Percentile: 85%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: spanish. Source rank: #42. Votes: 2957. Organization: google. License: Proprietary.

85% percentile inside its fair comparison set

1,450Raw benchmark valueCI 1,437 - 1,462

Text Arena · Chinese · No Style Control

AR · Multilingual · Human

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

Rank #13 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,506
Percentile: 95.9%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: chinese. Source rank: #18. Votes: 6483. Organization: google. License: Proprietary.

95.9% percentile inside its fair comparison set

1,506Raw benchmark valueCI 1,498 - 1,515

Text Arena · French · No Style Control

AR · Multilingual · Human

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

Rank #19 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,472
Percentile: 91.7%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: french. Source rank: #23. Votes: 2202. Organization: google. License: Proprietary.

91.7% percentile inside its fair comparison set

1,472Raw benchmark valueCI 1,458 - 1,487

Text Arena · German · No Style Control

AR · Multilingual · Human

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

Rank #7 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,485
Percentile: 97.5%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: german. Source rank: #8. Votes: 2500. Organization: google. License: Proprietary.

97.5% percentile inside its fair comparison set

1,485Raw benchmark valueCI 1,472 - 1,499

Text Arena · Japanese · No Style Control

AR · Multilingual · Human

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

Rank #9 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,466
Percentile: 96.1%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: japanese. Source rank: #12. Votes: 1817. Organization: google. License: Proprietary.

96.1% percentile inside its fair comparison set

1,466Raw benchmark valueCI 1,450 - 1,482

Text Arena · Korean · No Style Control

AR · Multilingual · Human

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

Rank #12 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,434
Percentile: 94.7%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: korean. Source rank: #15. Votes: 2399. Organization: google. License: Proprietary.

94.7% percentile inside its fair comparison set

1,434Raw benchmark valueCI 1,421 - 1,448

Text Arena · Russian · No Style Control

AR · Multilingual · Human

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

Rank #19 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,460
Percentile: 93.8%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: russian. Source rank: #24. Votes: 10032. Organization: google. License: Proprietary.

93.8% percentile inside its fair comparison set

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

Text Arena · Spanish · No Style Control

AR · Multilingual · Human

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

Rank #9 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,468
Percentile: 96.3%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: spanish. Source rank: #11. Votes: 2957. Organization: google. License: Proprietary.

96.3% percentile inside its fair comparison set

1,468Raw benchmark valueCI 1,456 - 1,481

Vision Arena · Chinese

AR · Multilingual · Human

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

Rank #31 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,266
Percentile: 61%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: chinese. Source rank: #40. Votes: 3057. Organization: google. License: Proprietary.

61% percentile inside its fair comparison set

1,266Raw benchmark valueCI 1,249 - 1,283

Vision Arena · Chinese · No Style Control

AR · Multilingual · Human

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

Rank #27 · Source label: gemini-2.5-pro

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,290
Percentile: 66.2%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `gemini-2.5-pro`. Category: chinese. Source rank: #34. Votes: 3057. Organization: google. License: Proprietary.

66.2% percentile inside its fair comparison set

1,290Raw benchmark valueCI 1,273 - 1,306