Model profile · Google

Gemini 3.1 Pro Preview

Closed weightsfrontier · registry tag 2026 preview

Thin verified coverage

Reads as thin verified coverage across the resolved source data.

Visible coverage: 41.1%
Verified coverage: 41.1%
Spread: n/a
Last verified: Aug 20, 2026

textcodevisiondocumentaudio5 aliases47 official source links

Open compare

Data version

Current snapshot.

Published Aug 20, 2026Model list checked2 stale sources216 claim warnings9 providers · 1154 tracked models

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

Source-linked scores by benchmark

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

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

Chat / text28 benchmarks82.1%

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 #23 · Source label: Gemini 3.1 Pro Preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 48
Percentile: 95.3%
Last updated: recent
Eligibility: preview_model

Parsed from Artificial Analysis public leaderboard field `intelligenceIndex`.

95.3% percentile inside its fair comparison set

48Raw 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 #8 · Source label: Gemini 3.1 Pro Preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 54.9%
Percentile: 98.1%
Last updated: recent
Eligibility: preview_model

Parsed from Artificial Analysis public leaderboard field `omniscienceAccuracy`.

98.1% percentile inside its fair comparison set

54.9%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 #56 · Source label: Gemini 3.1 Pro Preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 49.1%
Percentile: 85%
Last updated: recent
Eligibility: preview_model

Parsed from Artificial Analysis public leaderboard field `omniscienceNonHallucination`.

85% percentile inside its fair comparison set

49.1%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 #8 · Source label: Gemini 3.1 Pro Preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 77.1%
Percentile: 98%
Last updated: recent
Eligibility: preview_model

Parsed from Artificial Analysis public leaderboard field `ifbench`.

98% percentile inside its fair comparison set

77.1%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 #260 · Source label: Gemini 3.1 Pro Preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: $4.5 /1M tokens
Percentile: 14.6%
Last updated: recent
Eligibility: preview_model

Parsed from Artificial Analysis public leaderboard field `price1mBlended0To3To1`.

14.6% percentile inside its fair comparison set

$4.5 /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 #257 · Source label: Gemini 3.1 Pro Preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: $2 /1M input tokens
Percentile: 20.6%
Last updated: recent
Eligibility: preview_model

Parsed from Artificial Analysis public leaderboard field `price1mInputTokens`.

20.6% percentile inside its fair comparison set

$2 /1M input tokensRaw benchmark value

Output price

AA · Chat / text · Speed / cost

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

Rank #261 · Source label: Gemini 3.1 Pro Preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: $12 /1M output tokens
Percentile: 14.6%
Last updated: recent
Eligibility: preview_model

Parsed from Artificial Analysis public leaderboard field `price1mOutputTokens`.

14.6% percentile inside its fair comparison set

$12 /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 #70 · Source label: Gemini 3.1 Pro Preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 123.2 tokens/s
Percentile: 69.2%
Last updated: recent
Eligibility: preview_model

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

69.2% percentile inside its fair comparison set

123.2 tokens/sRaw benchmark value

Time to first token

AA · Chat / text · Speed / cost

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

Rank #205 · Source label: Gemini 3.1 Pro Preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 32.45s
Percentile: 8.9%
Last updated: recent
Eligibility: preview_model

Parsed from Artificial Analysis public leaderboard field `medianTimeToFirstTokenSeconds`.

8.9% percentile inside its fair comparison set

32.45sRaw 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 #167 · Source label: Gemini 3.1 Pro Preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 32.45s
Percentile: 25.9%
Last updated: recent
Eligibility: preview_model

Parsed from Artificial Analysis public leaderboard field `medianTimeToFirstAnswerTokenSeconds`.

25.9% percentile inside its fair comparison set

32.45sRaw 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 #5 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,486
Percentile: 98.8%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: overall. Source rank: #7. Votes: 60640. Organization: google. License: Proprietary.

98.8% percentile inside its fair comparison set

1,486Raw benchmark valueCI 1,483 - 1,490

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 #5 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,481
Percentile: 98.8%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: creative_writing. Source rank: #6. Votes: 9929. Organization: google. License: Proprietary.

98.8% percentile inside its fair comparison set

1,481Raw benchmark valueCI 1,474 - 1,488

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 #7 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,488
Percentile: 98.2%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: english. Source rank: #9. Votes: 28707. Organization: google. License: Proprietary.

98.2% percentile inside its fair comparison set

1,488Raw benchmark valueCI 1,483 - 1,493

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 #5 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,501
Percentile: 98.8%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: exclude_ties. Source rank: #7. Votes: 45610. Organization: google. License: Proprietary.

98.8% percentile inside its fair comparison set

1,501Raw benchmark valueCI 1,496 - 1,506

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 #6 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,508
Percentile: 98.5%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: hard_prompts. Source rank: #8. Votes: 38343. Organization: google. License: Proprietary.

98.5% percentile inside its fair comparison set

1,508Raw benchmark valueCI 1,503 - 1,512

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 #9 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,506
Percentile: 97.5%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: hard_prompts_english. Source rank: #11. Votes: 19092. Organization: google. License: Proprietary.

97.5% percentile inside its fair comparison set

1,506Raw benchmark valueCI 1,500 - 1,512

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 #7 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,482
Percentile: 98.2%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: instruction_following. Source rank: #9. Votes: 19527. Organization: google. License: Proprietary.

98.2% percentile inside its fair comparison set

1,482Raw benchmark valueCI 1,477 - 1,488

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 #6 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,500
Percentile: 98.4%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: longer_query. Source rank: #8. Votes: 23999. Organization: google. License: Proprietary.

98.4% percentile inside its fair comparison set

1,500Raw benchmark valueCI 1,494 - 1,505

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 #8 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,495
Percentile: 97.8%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: multi_turn. Source rank: #10. Votes: 10500. Organization: google. License: Proprietary.

97.8% percentile inside its fair comparison set

1,495Raw benchmark valueCI 1,488 - 1,501

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 #4 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,480
Percentile: 99.1%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: overall. Source rank: #6. Votes: 60640. Organization: google. License: Proprietary.

99.1% percentile inside its fair comparison set

1,480Raw benchmark valueCI 1,476 - 1,484

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 #3 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,483
Percentile: 99.4%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: creative_writing. Source rank: #4. Votes: 9929. Organization: google. License: Proprietary.

99.4% percentile inside its fair comparison set

1,483Raw benchmark valueCI 1,475 - 1,490

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 #10 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,477
Percentile: 97.2%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: english. Source rank: #12. Votes: 28707. Organization: google. License: Proprietary.

97.2% percentile inside its fair comparison set

1,477Raw benchmark valueCI 1,473 - 1,482

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 #5 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,490
Percentile: 98.8%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: exclude_ties. Source rank: #6. Votes: 45610. Organization: google. License: Proprietary.

98.8% percentile inside its fair comparison set

1,490Raw benchmark valueCI 1,485 - 1,495

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 #6 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,485
Percentile: 98.5%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: hard_prompts. Source rank: #8. Votes: 38343. Organization: google. License: Proprietary.

98.5% percentile inside its fair comparison set

1,485Raw benchmark valueCI 1,481 - 1,490

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 #13 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,480
Percentile: 96.3%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: hard_prompts_english. Source rank: #16. Votes: 19092. Organization: google. License: Proprietary.

96.3% percentile inside its fair comparison set

1,480Raw benchmark valueCI 1,475 - 1,486

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 #11 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,468
Percentile: 96.9%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: instruction_following. Source rank: #15. Votes: 19527. Organization: google. License: Proprietary.

96.9% percentile inside its fair comparison set

1,468Raw benchmark valueCI 1,463 - 1,474

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 #6 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,484
Percentile: 98.4%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: longer_query. Source rank: #9. Votes: 23999. Organization: google. License: Proprietary.

98.4% percentile inside its fair comparison set

1,484Raw benchmark valueCI 1,479 - 1,490

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 #6 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,486
Percentile: 98.5%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: multi_turn. Source rank: #8. Votes: 10500. Organization: google. License: Proprietary.

98.5% percentile inside its fair comparison set

1,486Raw benchmark valueCI 1,479 - 1,492

Coding24 benchmarks72.4%

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 #4 · Source label: Gemini 3.1 Pro Preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 53.8%
Percentile: 99.1%
Last updated: recent
Eligibility: preview_model

Parsed from Artificial Analysis public leaderboard field `terminalbenchHard`.

99.1% percentile inside its fair comparison set

53.8%Raw benchmark value

SciCode

AA · Coding · Objective

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

Rank #2 · Source label: Gemini 3.1 Pro Preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 58.9%
Percentile: 99.8%
Last updated: recent
Eligibility: preview_model

Parsed from Artificial Analysis public leaderboard field `scicode`.

99.8% percentile inside its fair comparison set

58.9%Raw benchmark value

Coding Index

AA · Coding · Combined

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

Rank #24 · Source label: Gemini 3.1 Pro Preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 69
Percentile: 87.8%
Last updated: recent
Eligibility: preview_model

Parsed from Artificial Analysis public leaderboard field `codingIndex`.

87.8% percentile inside its fair comparison set

69Raw benchmark value

Agentic Index

AA · Coding · Combined

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

Rank #63 · Source label: Gemini 3.1 Pro Preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 23
Percentile: 53.7%
Last updated: recent
Eligibility: preview_model

Parsed from Artificial Analysis public leaderboard field `agenticIndex`.

53.7% percentile inside its fair comparison set

23Raw 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 #21 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,447
Percentile: 72.6%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: overall. Source rank: #27. Votes: 12948. Organization: google. License: Proprietary.

72.6% percentile inside its fair comparison set

1,447Raw benchmark valueCI 1,441 - 1,453

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 #21 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,447
Percentile: 72.6%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: webdev. Source rank: #27. Votes: 12948. Organization: google. License: Proprietary.

72.6% percentile inside its fair comparison set

1,447Raw benchmark valueCI 1,441 - 1,453

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 #13 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,487
Percentile: 83.6%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: webdev-html. Source rank: #18. Votes: 1580. Organization: google. License: Proprietary.

83.6% percentile inside its fair comparison set

1,487Raw benchmark valueCI 1,471 - 1,502

Code Arena · Webdev React

AR · Coding · Human

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

Rank #22 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,435
Percentile: 64.4%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: webdev-react. Source rank: #28. Votes: 11333. Organization: google. License: Proprietary.

64.4% percentile inside its fair comparison set

1,435Raw benchmark valueCI 1,429 - 1,442

Code Migration

VALS-AI · Coding · Objective

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

Rank #15 · Source label: google/gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 17.3%
Percentile: 36.4%
Last updated: aging
Eligibility: preview_model

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

36.4% percentile inside its fair comparison set

17.3%Raw benchmark valueCI 9.6% - 25%

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 #2 · Source label: google/gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 88.5%
Percentile: 98.9%
Last updated: aging
Eligibility: preview_model

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

98.9% percentile inside its fair comparison set

88.5%Raw benchmark valueCI 86.7% - 90.3%

ProgramBench

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 #11 · Source label: google/gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 0%
Percentile: 70%
Last updated: aging
Eligibility: preview_model

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

70% percentile inside its fair comparison set

0%Raw benchmark valueCI 0% - 0%

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 #7 · Source label: google/gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 78.8%
Percentile: 90.7%
Last updated: aging
Eligibility: preview_model

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

90.7% percentile inside its fair comparison set

78.8%Raw benchmark valueCI 75.2% - 82.4%

Terminal-Bench 2.1

VALS-AI · Coding · Objective

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

Rank #5 · Source label: google/gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 70.8%
Percentile: 85.2%
Last updated: aging
Eligibility: preview_model

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

85.2% percentile inside its fair comparison set

70.8%Raw benchmark valueCI 68.6% - 73%

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 #21 · Source label: google/gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 32%
Percentile: 59.2%
Last updated: aging
Eligibility: preview_model

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

59.2% percentile inside its fair comparison set

32%Raw benchmark valueCI 23.5% - 40.5%

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 #11 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,523
Percentile: 96.9%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: coding. Source rank: #14. Votes: 16359. Organization: google. License: Proprietary.

96.9% percentile inside its fair comparison set

1,523Raw benchmark valueCI 1,517 - 1,529

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 #16 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,486
Percentile: 95.3%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: coding. Source rank: #20. Votes: 16359. Organization: google. License: Proprietary.

95.3% percentile inside its fair comparison set

1,486Raw benchmark valueCI 1,480 - 1,492

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 #7 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,490
Percentile: 64.7%
Last updated: stale
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: image_to_webdev. Source rank: #9. Votes: 3597. Organization: google. License: Proprietary.

64.7% percentile inside its fair comparison set

1,490Raw benchmark valueCI 1,478 - 1,502

Debugging

BB · Coding · Rubric

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

Rank #7 · Source label: google/gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: BridgeBench
Raw value: 85.9%
Percentile: 75%
Last updated: stale
Eligibility: preview_model

Parsed from the BridgeBench page for bridgebench-debugging.

75% percentile inside its fair comparison set

85.9%Raw benchmark value

Refactoring

BB · Coding · Rubric

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

Rank #3 · Source label: google/gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: BridgeBench
Raw value: 70%
Percentile: 90.5%
Last updated: stale
Eligibility: preview_model

Parsed from the BridgeBench page for bridgebench-refactoring.

90.5% percentile inside its fair comparison set

70%Raw benchmark value

Security

BB · Coding · Rubric

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

Rank #3 · Source label: google/gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: BridgeBench
Raw value: 85.2%
Percentile: 92%
Last updated: stale
Eligibility: preview_model

Parsed from the BridgeBench page for bridgebench-security.

92% percentile inside its fair comparison set

85.2%Raw benchmark value

UI bench

BB · Coding · Rubric

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

Rank #19 · Source label: google/gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: BridgeBench
Raw value: 52.9%
Percentile: 18.2%
Last updated: stale
Eligibility: preview_model

Parsed from the BridgeBench page for bridgebench-ui.

18.2% percentile inside its fair comparison set

52.9%Raw benchmark value

Speed throughput

BB · Coding · Speed / cost

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

Rank #10 · Source label: google/gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: BridgeBench
Raw value: 122.2 t/s
Percentile: 64%
Last updated: stale
Eligibility: preview_model

Parsed from the BridgeBench page for bridgebench-throughput.

64% percentile inside its fair comparison set

122.2 t/sRaw benchmark value

Speed TTFT

BB · Coding · Speed / cost

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

Rank #21 · Source label: google/gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: BridgeBench
Raw value: 7608.00ms
Percentile: 20%
Last updated: stale
Eligibility: preview_model

Parsed from the BridgeBench page for bridgebench-ttft.

20% percentile inside its fair comparison set

7608.00msRaw benchmark value

Speed cost

BB · Coding · Speed / cost

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

Rank #15 · Source label: google/gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: BridgeBench
Raw value: $0.1 $
Percentile: 48%
Last updated: stale
Eligibility: preview_model

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

48% percentile inside its fair comparison set

$0.1 $Raw benchmark value

Reasoning / math / science9 benchmarks85.6%

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 #8 · Source label: Gemini 3.1 Pro Preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 47%
Percentile: 98.4%
Last updated: recent
Eligibility: preview_model

Parsed from Artificial Analysis public leaderboard field `hle`.

98.4% percentile inside its fair comparison set

47%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 #3 · Source label: Gemini 3.1 Pro Preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 94.1%
Percentile: 99.6%
Last updated: recent
Eligibility: preview_model

Parsed from Artificial Analysis public leaderboard field `gpqa`.

99.6% percentile inside its fair comparison set

94.1%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 #16 · Source label: Gemini 3.1 Pro Preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 17.7%
Percentile: 96.2%
Last updated: recent
Eligibility: preview_model

Parsed from Artificial Analysis public leaderboard field `critpt`.

96.2% percentile inside its fair comparison set

17.7%Raw benchmark value

ProofBench

VALS-AI · Reasoning / math / science · Objective

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

Rank #12 · Source label: google/gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 26%
Percentile: 71.4%
Last updated: aging
Eligibility: preview_model

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

71.4% percentile inside its fair comparison set

26%Raw benchmark valueCI 17.4% - 34.6%

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 #1 · Source label: google/gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 95.5%
Percentile: 100%
Last updated: aging
Eligibility: preview_model

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

100% percentile inside its fair comparison set

95.5%Raw benchmark valueCI 93.4% - 97.5%

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 #2 · Source label: google/gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 91%
Percentile: 98.9%
Last updated: aging
Eligibility: preview_model

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

98.9% percentile inside its fair comparison set

91%Raw benchmark valueCI 90.4% - 91.5%

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 #8 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,495
Percentile: 97.8%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: math. Source rank: #10. Votes: 3429. Organization: google. License: Proprietary.

97.8% percentile inside its fair comparison set

1,495Raw benchmark valueCI 1,484 - 1,506

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 #6 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,491
Percentile: 98.4%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: math. Source rank: #9. Votes: 3429. Organization: google. License: Proprietary.

98.4% percentile inside its fair comparison set

1,491Raw benchmark valueCI 1,481 - 1,502

Reasoning

BB · Reasoning / math / science · Rubric

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

Rank #20 · Source label: google/gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: BridgeBench
Raw value: 34.3%
Percentile: 9.5%
Last updated: stale
Eligibility: preview_model

Parsed from the BridgeBench page for bridgebench-reasoning.

9.5% percentile inside its fair comparison set

34.3%Raw benchmark value

Professional reasoning32 benchmarks79.9%

APEX-Agents-AA

AA · Professional reasoning · Objective

Long-horizon agentic task completion.

Rank #9 · Source label: Gemini 3.1 Pro Preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 32%
Percentile: 71.4%
Last updated: recent
Eligibility: preview_model

Parsed from Artificial Analysis public leaderboard field `apexAgents`.

71.4% percentile inside its fair comparison set

32%Raw benchmark value

GDPval-AA

AA · Professional reasoning · Rubric

Agentic performance on economically valuable work tasks.

Rank #79 · Source label: Gemini 3.1 Pro Preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 965
Percentile: 54.1%
Last updated: recent
Eligibility: preview_model

Parsed from Artificial Analysis public leaderboard field `gdpvalBreakdown.elo`.

54.1% percentile inside its fair comparison set

965Raw benchmark value

Legal Research Bench

VALS-AI · Professional reasoning · Objective

Applied legal research tasks.

Rank #10 · Source label: google/gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 20.7%
Percentile: 25%
Last updated: aging
Eligibility: preview_model

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

25% percentile inside its fair comparison set

20.7%Raw benchmark valueCI 15.2% - 26.2%

Public Benefits Bench

VALS-AI · Professional reasoning · Objective

Answering SNAP benefits questions across the public-benefits lifecycle.

Rank #10 · Source label: google/gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 53.8%
Percentile: 18.2%
Last updated: aging
Eligibility: preview_model

Parsed from Vals AI BenchmarkView overall scores. Vals slug: public-benefits-bench; provider: Google.

18.2% percentile inside its fair comparison set

53.8%Raw benchmark valueCI 51.2% - 56.3%

Vals Index

VALS-AI · Professional reasoning · Combined

Weighted model performance across economically relevant Vals tasks.

Rank #12 · Source label: google/gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 54
Percentile: 57.7%
Last updated: aging
Eligibility: preview_model

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

57.7% percentile inside its fair comparison set

54Raw benchmark valueCI 51 - 57

LegalBench

VALS-AI · Professional reasoning · Objective

Academic legal reasoning tasks.

Rank #2 · Source label: google/gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 87.4%
Percentile: 98.9%
Last updated: aging
Eligibility: preview_model

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

98.9% percentile inside its fair comparison set

87.4%Raw benchmark valueCI 86.8% - 88%

Finance Agent v2

VALS-AI · Professional reasoning · Objective

Core financial analyst tasks for agentic models.

Rank #13 · Source label: google/gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 43%
Percentile: 52%
Last updated: aging
Eligibility: preview_model

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

52% percentile inside its fair comparison set

43%Raw benchmark valueCI 40.6% - 45.3%

TaxEval v2

VALS-AI · Professional reasoning · Objective

Answer quality on tax questions and responses.

Rank #39 · Source label: google/gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 72.9%
Percentile: 58.2%
Last updated: aging
Eligibility: preview_model

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

58.2% 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 #1 · Source label: google/gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 59.1%
Percentile: 100%
Last updated: aging
Eligibility: preview_model

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

100% percentile inside its fair comparison set

59.1%Raw benchmark valueCI 55.1% - 63%

MedScribe

VALS-AI · Professional reasoning · Objective

Administrative documentation support for doctors.

Rank #31 · Source label: google/gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 76.1%
Percentile: 40%
Last updated: aging
Eligibility: preview_model

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

40% percentile inside its fair comparison set

76.1%Raw benchmark valueCI 72.4% - 79.9%

Text Arena · Expert

AR · Professional reasoning · Human

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

Rank #8 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,515
Percentile: 97.5%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: expert. Source rank: #10. Votes: 5352. Organization: google. License: Proprietary.

97.5% percentile inside its fair comparison set

1,515Raw benchmark valueCI 1,506 - 1,524

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 #11 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,477
Percentile: 96.9%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: industry_business_and_management_and_financial_operations. Source rank: #14. Votes: 11754. Organization: google. License: Proprietary.

96.9% percentile inside its fair comparison set

1,477Raw benchmark valueCI 1,471 - 1,484

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 #5 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,467
Percentile: 98.8%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: industry_entertainment_and_sports_and_media. Source rank: #7. Votes: 12538. Organization: google. License: Proprietary.

98.8% percentile inside its fair comparison set

1,467Raw benchmark valueCI 1,461 - 1,474

Text Arena · Industry Legal And Government

AR · Professional reasoning · Human

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

Rank #8 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,495
Percentile: 97.7%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: industry_legal_and_government. Source rank: #9. Votes: 4683. Organization: google. License: Proprietary.

97.7% percentile inside its fair comparison set

1,495Raw benchmark valueCI 1,486 - 1,505

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 #6 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,508
Percentile: 98.5%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: industry_life_and_physical_and_social_science. Source rank: #8. Votes: 9896. Organization: google. License: Proprietary.

98.5% percentile inside its fair comparison set

1,508Raw benchmark valueCI 1,501 - 1,515

Text Arena · Industry Mathematical

AR · Professional reasoning · Human

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

Rank #12 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,490
Percentile: 96.4%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: industry_mathematical. Source rank: #15. Votes: 3217. Organization: google. License: Proprietary.

96.4% percentile inside its fair comparison set

1,490Raw benchmark valueCI 1,479 - 1,501

Text Arena · Industry Medicine And Healthcare

AR · Professional reasoning · Human

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

Rank #6 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,505
Percentile: 98.3%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: industry_medicine_and_healthcare. Source rank: #8. Votes: 4474. Organization: google. License: Proprietary.

98.3% percentile inside its fair comparison set

1,505Raw benchmark valueCI 1,495 - 1,514

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 #9 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,514
Percentile: 97.5%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: industry_software_and_it_services. Source rank: #12. Votes: 23570. Organization: google. License: Proprietary.

97.5% percentile inside its fair comparison set

1,514Raw benchmark valueCI 1,509 - 1,519

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 #5 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,479
Percentile: 98.8%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: industry_writing_and_literature_and_language. Source rank: #7. Votes: 14515. Organization: google. License: Proprietary.

98.8% percentile inside its fair comparison set

1,479Raw benchmark valueCI 1,473 - 1,485

Text Arena · Expert · No Style Control

AR · Professional reasoning · Human

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

Rank #12 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,492
Percentile: 96%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: expert. Source rank: #15. Votes: 5352. Organization: google. License: Proprietary.

96% percentile inside its fair comparison set

1,492Raw benchmark valueCI 1,483 - 1,501

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 #12 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,461
Percentile: 96.5%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: industry_business_and_management_and_financial_operations. Source rank: #15. Votes: 11754. Organization: google. License: Proprietary.

96.5% percentile inside its fair comparison set

1,461Raw benchmark valueCI 1,454 - 1,467

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 #4 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,465
Percentile: 99.1%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: industry_entertainment_and_sports_and_media. Source rank: #6. Votes: 12538. Organization: google. License: Proprietary.

99.1% percentile inside its fair comparison set

1,465Raw benchmark valueCI 1,458 - 1,471

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 #6 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,485
Percentile: 98.3%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: industry_legal_and_government. Source rank: #7. Votes: 4683. Organization: google. License: Proprietary.

98.3% percentile inside its fair comparison set

1,485Raw benchmark valueCI 1,476 - 1,495

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 #6 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,497
Percentile: 98.5%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: industry_life_and_physical_and_social_science. Source rank: #7. Votes: 9896. Organization: google. License: Proprietary.

98.5% percentile inside its fair comparison set

1,497Raw benchmark valueCI 1,490 - 1,503

Text Arena · Industry Mathematical · No Style Control

AR · Professional reasoning · Human

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

Rank #14 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,481
Percentile: 95.8%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: industry_mathematical. Source rank: #16. Votes: 3217. Organization: google. License: Proprietary.

95.8% percentile inside its fair comparison set

1,481Raw benchmark valueCI 1,470 - 1,493

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 #6 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,488
Percentile: 98.3%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: industry_medicine_and_healthcare. Source rank: #8. Votes: 4474. Organization: google. License: Proprietary.

98.3% percentile inside its fair comparison set

1,488Raw benchmark valueCI 1,478 - 1,498

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 #13 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,488
Percentile: 96.3%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: industry_software_and_it_services. Source rank: #16. Votes: 23570. Organization: google. License: Proprietary.

96.3% percentile inside its fair comparison set

1,488Raw benchmark valueCI 1,483 - 1,493

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 #4 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,479
Percentile: 99.1%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: industry_writing_and_literature_and_language. Source rank: #5. Votes: 14515. Organization: google. License: Proprietary.

99.1% percentile inside its fair comparison set

1,479Raw benchmark valueCI 1,473 - 1,486

SAGE

VALS-AI · Professional reasoning · Objective

Student Assessment with Generative Evaluation.

Rank #16 · Source label: google/gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 48.7%
Percentile: 66.7%
Last updated: aging
Eligibility: preview_model

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

66.7% percentile inside its fair comparison set

48.7%Raw benchmark valueCI 42.2% - 55.1%

Public Benefits Bench v1

VALS-AI · Professional reasoning · Objective

Answering public-benefits questions across the benefits lifecycle.

Rank #10 · Source label: google/gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 50.9%
Percentile: 25%
Last updated: aging
Eligibility: preview_model

Parsed from Vals AI BenchmarkView overall scores. Vals slug: public-benefits-bench-v1; provider: Google.

25% percentile inside its fair comparison set

50.9%Raw benchmark valueCI 48.3% - 53.4%

Hallucination

BB · Professional reasoning · Rubric

Accuracy and fabrication resistance under prompts that invite unsupported claims.

Rank #2 · Source label: google/gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: BridgeBench
Raw value: 79.1%
Percentile: 97.1%
Last updated: stale
Eligibility: preview_model

Parsed from the BridgeBench page for bridgebench-hallucination.

97.1% percentile inside its fair comparison set

79.1%Raw benchmark value

BS pushback

BB · Professional reasoning · Rubric

Resistance to confidently accepting bogus assumptions in expert-style prompts.

Rank #13 · Source label: google/gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: BridgeBench
Raw value: 66.5%
Percentile: 33.3%
Last updated: stale
Eligibility: preview_model

Parsed from the BridgeBench page for bridgebench-pushback.

33.3% percentile inside its fair comparison set

66.5%Raw benchmark value

Search / tool use1 benchmark96.4%

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 #13 · Source label: Gemini 3.1 Pro Preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 95.6%
Percentile: 96.4%
Last updated: recent
Eligibility: preview_model

Parsed from Artificial Analysis public leaderboard field `tau2`.

96.4% percentile inside its fair comparison set

95.6%Raw benchmark value

Long context2 benchmarks85.5%

Long Context Reasoning

AA · Long context · Objective

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

Rank #9 · Source label: Gemini 3.1 Pro Preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 79%
Percentile: 98.2%
Last updated: recent
Eligibility: preview_model

Parsed from Artificial Analysis public leaderboard field `lcr`.

98.2% percentile inside its fair comparison set

79%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 #25 · Source label: google/gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 64.5%
Percentile: 72.7%
Last updated: aging
Eligibility: preview_model

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

72.7% percentile inside its fair comparison set

64.5%Raw benchmark valueCI 62.6% - 66.3%

Vision understanding20 benchmarks92.2%

MMMU-Pro

AA · Vision understanding · Objective

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

Rank #8 · Source label: Gemini 3.1 Pro Preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 82.4%
Percentile: 96.3%
Last updated: recent
Eligibility: preview_model

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

96.3% percentile inside its fair comparison set

82.4%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 #10 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,280
Percentile: 91.7%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: overall. Source rank: #13. Votes: 26478. Organization: google. License: Proprietary.

91.7% percentile inside its fair comparison set

1,280Raw benchmark valueCI 1,273 - 1,286

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 #3 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,258
Percentile: 92.3%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: captioning. Source rank: #3. Votes: 97. Organization: google. License: Proprietary.

92.3% percentile inside its fair comparison set

1,258Raw benchmark valueCI 1,205 - 1,311

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 #5 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,294
Percentile: 92.7%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: creative_writing_vision. Source rank: #6. Votes: 1507. Organization: google. License: Proprietary.

92.7% percentile inside its fair comparison set

1,294Raw benchmark valueCI 1,278 - 1,311

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 #8 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,311
Percentile: 90%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: diagram. Source rank: #11. Votes: 6941. Organization: google. License: Proprietary.

90% percentile inside its fair comparison set

1,311Raw benchmark valueCI 1,302 - 1,320

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 #14 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,272
Percentile: 88.1%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: english. Source rank: #17. Votes: 11106. Organization: google. License: Proprietary.

88.1% percentile inside its fair comparison set

1,272Raw benchmark valueCI 1,263 - 1,281

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 #3 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,293
Percentile: 93.8%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: entity_recognition. Source rank: #3. Votes: 218. Organization: google. License: Proprietary.

93.8% percentile inside its fair comparison set

1,293Raw benchmark valueCI 1,255 - 1,331

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 #8 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,311
Percentile: 89.7%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: homework. Source rank: #11. Votes: 4011. Organization: google. License: Proprietary.

89.7% percentile inside its fair comparison set

1,311Raw benchmark valueCI 1,301 - 1,322

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 #5 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,300
Percentile: 91.8%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: humor. Source rank: #5. Votes: 994. Organization: google. License: Proprietary.

91.8% percentile inside its fair comparison set

1,300Raw benchmark valueCI 1,280 - 1,319

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 #9 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,297
Percentile: 88.6%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: ocr. Source rank: #11. Votes: 18511. Organization: google. License: Proprietary.

88.6% percentile inside its fair comparison set

1,297Raw benchmark valueCI 1,291 - 1,303

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 #7 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,297
Percentile: 94.5%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: overall. Source rank: #9. Votes: 26478. Organization: google. License: Proprietary.

94.5% percentile inside its fair comparison set

1,297Raw benchmark valueCI 1,291 - 1,303

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 #2 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,316
Percentile: 96.2%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: captioning. Source rank: #2. Votes: 97. Organization: google. License: Proprietary.

96.2% percentile inside its fair comparison set

1,316Raw benchmark valueCI 1,263 - 1,369

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 #5 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,317
Percentile: 92.7%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: creative_writing_vision. Source rank: #6. Votes: 1507. Organization: google. License: Proprietary.

92.7% percentile inside its fair comparison set

1,317Raw benchmark valueCI 1,301 - 1,333

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 #6 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,315
Percentile: 92.9%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: diagram. Source rank: #9. Votes: 6941. Organization: google. License: Proprietary.

92.9% percentile inside its fair comparison set

1,315Raw benchmark valueCI 1,306 - 1,324

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 #12 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,288
Percentile: 89.9%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: english. Source rank: #14. Votes: 11106. Organization: google. License: Proprietary.

89.9% percentile inside its fair comparison set

1,288Raw benchmark valueCI 1,279 - 1,296

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 #3 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,335
Percentile: 93.8%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: entity_recognition. Source rank: #3. Votes: 218. Organization: google. License: Proprietary.

93.8% percentile inside its fair comparison set

1,335Raw benchmark valueCI 1,297 - 1,373

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 #7 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,318
Percentile: 91.2%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: homework. Source rank: #10. Votes: 4011. Organization: google. License: Proprietary.

91.2% percentile inside its fair comparison set

1,318Raw benchmark valueCI 1,308 - 1,328

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 #4 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,323
Percentile: 93.9%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: humor. Source rank: #4. Votes: 994. Organization: google. License: Proprietary.

93.9% percentile inside its fair comparison set

1,323Raw benchmark valueCI 1,303 - 1,343

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 #8 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,308
Percentile: 90%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: ocr. Source rank: #10. Votes: 18511. Organization: google. License: Proprietary.

90% percentile inside its fair comparison set

1,308Raw benchmark valueCI 1,302 - 1,314

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 #4 · Source label: google/gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 88.2%
Percentile: 94.8%
Last updated: aging
Eligibility: preview_model

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

94.8% percentile inside its fair comparison set

88.2%Raw benchmark valueCI 86.7% - 89.7%

Document understanding4 benchmarks64.7%

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 #12 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,455
Percentile: 54.2%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: overall. Source rank: #15. Votes: 38090. Organization: google. License: Proprietary.

54.2% percentile inside its fair comparison set

1,455Raw benchmark valueCI 1,450 - 1,461

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 #13 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,441
Percentile: 50%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: overall. Source rank: #16. Votes: 38090. Organization: google. License: Proprietary.

50% percentile inside its fair comparison set

1,441Raw benchmark valueCI 1,436 - 1,447

Vals Multimodal Index

VALS-AI · Document understanding · Combined

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

Rank #9 · Source label: google/gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 56
Percentile: 57.9%
Last updated: aging
Eligibility: preview_model

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

57.9% percentile inside its fair comparison set

56Raw benchmark valueCI 53 - 59

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 #3 · Source label: google/gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 69.4%
Percentile: 96.7%
Last updated: aging
Eligibility: preview_model

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

96.7% percentile inside its fair comparison set

69.4%Raw benchmark valueCI 67.6% - 71.2%

Multilingual16 benchmarks98.1%

Text Arena · Chinese

AR · Multilingual · Human

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

Rank #7 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,528
Percentile: 98%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: chinese. Source rank: #9. Votes: 3203. Organization: google. License: Proprietary.

98% percentile inside its fair comparison set

1,528Raw benchmark valueCI 1,516 - 1,539

Text Arena · French

AR · Multilingual · Human

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

Rank #9 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,505
Percentile: 96.3%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: french. Source rank: #10. Votes: 1967. Organization: google. License: Proprietary.

96.3% percentile inside its fair comparison set

1,505Raw benchmark valueCI 1,489 - 1,521

Text Arena · German

AR · Multilingual · Human

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

Rank #5 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,497
Percentile: 98.3%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: german. Source rank: #5. Votes: 1012. Organization: google. License: Proprietary.

98.3% percentile inside its fair comparison set

1,497Raw benchmark valueCI 1,477 - 1,517

Text Arena · Japanese

AR · Multilingual · Human

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

Rank #2 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,510
Percentile: 99.5%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: japanese. Source rank: #2. Votes: 535. Organization: google. License: Proprietary.

99.5% percentile inside its fair comparison set

1,510Raw benchmark valueCI 1,482 - 1,538

Text Arena · Korean

AR · Multilingual · Human

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

Rank #4 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,466
Percentile: 98.6%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: korean. Source rank: #4. Votes: 1054. Organization: google. License: Proprietary.

98.6% percentile inside its fair comparison set

1,466Raw benchmark valueCI 1,445 - 1,487

Text Arena · Russian

AR · Multilingual · Human

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

Rank #6 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,501
Percentile: 98.3%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: russian. Source rank: #7. Votes: 6426. Organization: google. License: Proprietary.

98.3% percentile inside its fair comparison set

1,501Raw benchmark valueCI 1,492 - 1,509

Text Arena · Spanish

AR · Multilingual · Human

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

Rank #3 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,484
Percentile: 99.1%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: spanish. Source rank: #4. Votes: 1960. Organization: google. License: Proprietary.

99.1% percentile inside its fair comparison set

1,484Raw benchmark valueCI 1,468 - 1,499

Text Arena · Chinese · No Style Control

AR · Multilingual · Human

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

Rank #7 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,528
Percentile: 98%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: chinese. Source rank: #10. Votes: 3203. Organization: google. License: Proprietary.

98% percentile inside its fair comparison set

1,528Raw benchmark valueCI 1,516 - 1,539

Text Arena · French · No Style Control

AR · Multilingual · Human

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

Rank #12 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,490
Percentile: 94.9%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: french. Source rank: #14. Votes: 1967. Organization: google. License: Proprietary.

94.9% percentile inside its fair comparison set

1,490Raw benchmark valueCI 1,475 - 1,506

Text Arena · German · No Style Control

AR · Multilingual · Human

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

Rank #4 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,496
Percentile: 98.7%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: german. Source rank: #4. Votes: 1012. Organization: google. License: Proprietary.

98.7% percentile inside its fair comparison set

1,496Raw benchmark valueCI 1,476 - 1,515

Text Arena · Japanese · No Style Control

AR · Multilingual · Human

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

Rank #4 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,507
Percentile: 98.5%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: japanese. Source rank: #4. Votes: 535. Organization: google. License: Proprietary.

98.5% percentile inside its fair comparison set

1,507Raw benchmark valueCI 1,479 - 1,535

Text Arena · Korean · No Style Control

AR · Multilingual · Human

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

Rank #3 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,468
Percentile: 99%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: korean. Source rank: #3. Votes: 1054. Organization: google. License: Proprietary.

99% percentile inside its fair comparison set

1,468Raw benchmark valueCI 1,447 - 1,488

Text Arena · Russian · No Style Control

AR · Multilingual · Human

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

Rank #3 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,497
Percentile: 99.3%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: russian. Source rank: #3. Votes: 6426. Organization: google. License: Proprietary.

99.3% percentile inside its fair comparison set

1,497Raw benchmark valueCI 1,489 - 1,505

Text Arena · Spanish · No Style Control

AR · Multilingual · Human

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

Rank #2 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,483
Percentile: 99.5%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: spanish. Source rank: #3. Votes: 1960. Organization: google. License: Proprietary.

99.5% percentile inside its fair comparison set

1,483Raw benchmark valueCI 1,468 - 1,498

Vision Arena · Chinese

AR · Multilingual · Human

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

Rank #4 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,344
Percentile: 96.1%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: chinese. Source rank: #5. Votes: 1599. Organization: google. License: Proprietary.

96.1% percentile inside its fair comparison set

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

Vision Arena · Chinese · No Style Control

AR · Multilingual · Human

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

Rank #3 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,369
Percentile: 97.4%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: chinese. Source rank: #3. Votes: 1599. Organization: google. License: Proprietary.

97.4% percentile inside its fair comparison set

1,369Raw benchmark valueCI 1,349 - 1,389

Source links and registry checks

official

Google Gemini models docs

Aug 20, 2026

source →

official

Arena

Aug 20, 2026

source →

official

Arena

Aug 20, 2026

source →

official

Arena

Aug 20, 2026

source →

official

Arena

Aug 20, 2026

source →

official

Arena

Aug 20, 2026

source →

official

Arena

Aug 20, 2026

source →

official

Arena

Aug 20, 2026

source →

official

Arena

Aug 20, 2026

source →

official

Arena

Aug 20, 2026

source →

official

Arena

Aug 20, 2026

source →

official

Arena

Aug 20, 2026

source →

official

Arena

Aug 20, 2026

source →

official

Arena

Aug 20, 2026

source →

official

Arena

Aug 20, 2026

source →

official

Arena

Aug 20, 2026

source →

official

Arena

Aug 20, 2026

source →

official

Arena

Aug 20, 2026

source →

official

Arena

Aug 20, 2026

source →

official

Arena

Aug 20, 2026

source →

official

Arena

Aug 20, 2026

source →

official

Arena

Aug 20, 2026

source →

official

Arena

Aug 20, 2026

source →

official

Arena

Aug 20, 2026

source →

official

Arena

Aug 20, 2026

source →

official

Arena

Aug 20, 2026

source →

official

Arena

Aug 20, 2026

source →

official

Arena

Aug 20, 2026

source →

official

Arena

Aug 20, 2026

source →

official

Arena

Aug 20, 2026

source →

official

Arena

Aug 20, 2026

source →

official

Arena

Aug 20, 2026

source →

official

Arena

Aug 20, 2026

source →

official

Arena

Aug 20, 2026

source →

official

Arena

Aug 20, 2026

source →

official

Arena

Aug 20, 2026

source →

official

Arena

Aug 20, 2026

source →

official

Arena

Aug 20, 2026

source →

official

Arena

Aug 20, 2026

source →

official

Arena

Aug 20, 2026

source →

official

Arena

Aug 20, 2026

source →

official

Arena

Aug 20, 2026

source →

official

Arena

Aug 20, 2026

source →

official

Arena

Aug 20, 2026

source →

official

Arena

Aug 20, 2026

source →

official

Arena

Aug 20, 2026

source →

official

Artificial Analysis

Aug 20, 2026

source →

Model profile · Google

Gemini 3.1 Pro Preview

Closed weightsfrontier · registry tag 2026 preview

Thin verified coverage

Reads as thin verified coverage across the resolved source data.

Visible coverage: 41.1%
Verified coverage: 41.1%
Spread: n/a
Last verified: Aug 20, 2026

textcodevisiondocumentaudio5 aliases47 official source links

Open compare

Data version

Current snapshot.

Published Aug 20, 2026Model list checked2 stale sources216 claim warnings9 providers · 1154 tracked models

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

Source-linked scores by benchmark

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

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

Chat / text28 benchmarks82.1%

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 #23 · Source label: Gemini 3.1 Pro Preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 48
Percentile: 95.3%
Last updated: recent
Eligibility: preview_model

Parsed from Artificial Analysis public leaderboard field `intelligenceIndex`.

95.3% percentile inside its fair comparison set

48Raw 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 #8 · Source label: Gemini 3.1 Pro Preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 54.9%
Percentile: 98.1%
Last updated: recent
Eligibility: preview_model

Parsed from Artificial Analysis public leaderboard field `omniscienceAccuracy`.

98.1% percentile inside its fair comparison set

54.9%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 #56 · Source label: Gemini 3.1 Pro Preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 49.1%
Percentile: 85%
Last updated: recent
Eligibility: preview_model

Parsed from Artificial Analysis public leaderboard field `omniscienceNonHallucination`.

85% percentile inside its fair comparison set

49.1%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 #8 · Source label: Gemini 3.1 Pro Preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 77.1%
Percentile: 98%
Last updated: recent
Eligibility: preview_model

Parsed from Artificial Analysis public leaderboard field `ifbench`.

98% percentile inside its fair comparison set

77.1%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 #260 · Source label: Gemini 3.1 Pro Preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: $4.5 /1M tokens
Percentile: 14.6%
Last updated: recent
Eligibility: preview_model

Parsed from Artificial Analysis public leaderboard field `price1mBlended0To3To1`.

14.6% percentile inside its fair comparison set

$4.5 /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 #257 · Source label: Gemini 3.1 Pro Preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: $2 /1M input tokens
Percentile: 20.6%
Last updated: recent
Eligibility: preview_model

Parsed from Artificial Analysis public leaderboard field `price1mInputTokens`.

20.6% percentile inside its fair comparison set

$2 /1M input tokensRaw benchmark value

Output price

AA · Chat / text · Speed / cost

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

Rank #261 · Source label: Gemini 3.1 Pro Preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: $12 /1M output tokens
Percentile: 14.6%
Last updated: recent
Eligibility: preview_model

Parsed from Artificial Analysis public leaderboard field `price1mOutputTokens`.

14.6% percentile inside its fair comparison set

$12 /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 #70 · Source label: Gemini 3.1 Pro Preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 123.2 tokens/s
Percentile: 69.2%
Last updated: recent
Eligibility: preview_model

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

69.2% percentile inside its fair comparison set

123.2 tokens/sRaw benchmark value

Time to first token

AA · Chat / text · Speed / cost

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

Rank #205 · Source label: Gemini 3.1 Pro Preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 32.45s
Percentile: 8.9%
Last updated: recent
Eligibility: preview_model

Parsed from Artificial Analysis public leaderboard field `medianTimeToFirstTokenSeconds`.

8.9% percentile inside its fair comparison set

32.45sRaw 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 #167 · Source label: Gemini 3.1 Pro Preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 32.45s
Percentile: 25.9%
Last updated: recent
Eligibility: preview_model

Parsed from Artificial Analysis public leaderboard field `medianTimeToFirstAnswerTokenSeconds`.

25.9% percentile inside its fair comparison set

32.45sRaw 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 #5 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,486
Percentile: 98.8%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: overall. Source rank: #7. Votes: 60640. Organization: google. License: Proprietary.

98.8% percentile inside its fair comparison set

1,486Raw benchmark valueCI 1,483 - 1,490

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 #5 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,481
Percentile: 98.8%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: creative_writing. Source rank: #6. Votes: 9929. Organization: google. License: Proprietary.

98.8% percentile inside its fair comparison set

1,481Raw benchmark valueCI 1,474 - 1,488

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 #7 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,488
Percentile: 98.2%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: english. Source rank: #9. Votes: 28707. Organization: google. License: Proprietary.

98.2% percentile inside its fair comparison set

1,488Raw benchmark valueCI 1,483 - 1,493

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 #5 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,501
Percentile: 98.8%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: exclude_ties. Source rank: #7. Votes: 45610. Organization: google. License: Proprietary.

98.8% percentile inside its fair comparison set

1,501Raw benchmark valueCI 1,496 - 1,506

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 #6 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,508
Percentile: 98.5%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: hard_prompts. Source rank: #8. Votes: 38343. Organization: google. License: Proprietary.

98.5% percentile inside its fair comparison set

1,508Raw benchmark valueCI 1,503 - 1,512

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 #9 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,506
Percentile: 97.5%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: hard_prompts_english. Source rank: #11. Votes: 19092. Organization: google. License: Proprietary.

97.5% percentile inside its fair comparison set

1,506Raw benchmark valueCI 1,500 - 1,512

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 #7 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,482
Percentile: 98.2%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: instruction_following. Source rank: #9. Votes: 19527. Organization: google. License: Proprietary.

98.2% percentile inside its fair comparison set

1,482Raw benchmark valueCI 1,477 - 1,488

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 #6 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,500
Percentile: 98.4%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: longer_query. Source rank: #8. Votes: 23999. Organization: google. License: Proprietary.

98.4% percentile inside its fair comparison set

1,500Raw benchmark valueCI 1,494 - 1,505

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 #8 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,495
Percentile: 97.8%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: multi_turn. Source rank: #10. Votes: 10500. Organization: google. License: Proprietary.

97.8% percentile inside its fair comparison set

1,495Raw benchmark valueCI 1,488 - 1,501

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 #4 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,480
Percentile: 99.1%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: overall. Source rank: #6. Votes: 60640. Organization: google. License: Proprietary.

99.1% percentile inside its fair comparison set

1,480Raw benchmark valueCI 1,476 - 1,484

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 #3 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,483
Percentile: 99.4%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: creative_writing. Source rank: #4. Votes: 9929. Organization: google. License: Proprietary.

99.4% percentile inside its fair comparison set

1,483Raw benchmark valueCI 1,475 - 1,490

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 #10 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,477
Percentile: 97.2%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: english. Source rank: #12. Votes: 28707. Organization: google. License: Proprietary.

97.2% percentile inside its fair comparison set

1,477Raw benchmark valueCI 1,473 - 1,482

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 #5 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,490
Percentile: 98.8%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: exclude_ties. Source rank: #6. Votes: 45610. Organization: google. License: Proprietary.

98.8% percentile inside its fair comparison set

1,490Raw benchmark valueCI 1,485 - 1,495

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 #6 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,485
Percentile: 98.5%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: hard_prompts. Source rank: #8. Votes: 38343. Organization: google. License: Proprietary.

98.5% percentile inside its fair comparison set

1,485Raw benchmark valueCI 1,481 - 1,490

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 #13 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,480
Percentile: 96.3%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: hard_prompts_english. Source rank: #16. Votes: 19092. Organization: google. License: Proprietary.

96.3% percentile inside its fair comparison set

1,480Raw benchmark valueCI 1,475 - 1,486

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 #11 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,468
Percentile: 96.9%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: instruction_following. Source rank: #15. Votes: 19527. Organization: google. License: Proprietary.

96.9% percentile inside its fair comparison set

1,468Raw benchmark valueCI 1,463 - 1,474

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 #6 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,484
Percentile: 98.4%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: longer_query. Source rank: #9. Votes: 23999. Organization: google. License: Proprietary.

98.4% percentile inside its fair comparison set

1,484Raw benchmark valueCI 1,479 - 1,490

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 #6 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,486
Percentile: 98.5%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: multi_turn. Source rank: #8. Votes: 10500. Organization: google. License: Proprietary.

98.5% percentile inside its fair comparison set

1,486Raw benchmark valueCI 1,479 - 1,492

Coding24 benchmarks72.4%

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 #4 · Source label: Gemini 3.1 Pro Preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 53.8%
Percentile: 99.1%
Last updated: recent
Eligibility: preview_model

Parsed from Artificial Analysis public leaderboard field `terminalbenchHard`.

99.1% percentile inside its fair comparison set

53.8%Raw benchmark value

SciCode

AA · Coding · Objective

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

Rank #2 · Source label: Gemini 3.1 Pro Preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 58.9%
Percentile: 99.8%
Last updated: recent
Eligibility: preview_model

Parsed from Artificial Analysis public leaderboard field `scicode`.

99.8% percentile inside its fair comparison set

58.9%Raw benchmark value

Coding Index

AA · Coding · Combined

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

Rank #24 · Source label: Gemini 3.1 Pro Preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 69
Percentile: 87.8%
Last updated: recent
Eligibility: preview_model

Parsed from Artificial Analysis public leaderboard field `codingIndex`.

87.8% percentile inside its fair comparison set

69Raw benchmark value

Agentic Index

AA · Coding · Combined

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

Rank #63 · Source label: Gemini 3.1 Pro Preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 23
Percentile: 53.7%
Last updated: recent
Eligibility: preview_model

Parsed from Artificial Analysis public leaderboard field `agenticIndex`.

53.7% percentile inside its fair comparison set

23Raw 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 #21 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,447
Percentile: 72.6%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: overall. Source rank: #27. Votes: 12948. Organization: google. License: Proprietary.

72.6% percentile inside its fair comparison set

1,447Raw benchmark valueCI 1,441 - 1,453

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 #21 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,447
Percentile: 72.6%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: webdev. Source rank: #27. Votes: 12948. Organization: google. License: Proprietary.

72.6% percentile inside its fair comparison set

1,447Raw benchmark valueCI 1,441 - 1,453

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 #13 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,487
Percentile: 83.6%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: webdev-html. Source rank: #18. Votes: 1580. Organization: google. License: Proprietary.

83.6% percentile inside its fair comparison set

1,487Raw benchmark valueCI 1,471 - 1,502

Code Arena · Webdev React

AR · Coding · Human

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

Rank #22 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,435
Percentile: 64.4%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: webdev-react. Source rank: #28. Votes: 11333. Organization: google. License: Proprietary.

64.4% percentile inside its fair comparison set

1,435Raw benchmark valueCI 1,429 - 1,442

Code Migration

VALS-AI · Coding · Objective

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

Rank #15 · Source label: google/gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 17.3%
Percentile: 36.4%
Last updated: aging
Eligibility: preview_model

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

36.4% percentile inside its fair comparison set

17.3%Raw benchmark valueCI 9.6% - 25%

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 #2 · Source label: google/gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 88.5%
Percentile: 98.9%
Last updated: aging
Eligibility: preview_model

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

98.9% percentile inside its fair comparison set

88.5%Raw benchmark valueCI 86.7% - 90.3%

ProgramBench

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 #11 · Source label: google/gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 0%
Percentile: 70%
Last updated: aging
Eligibility: preview_model

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

70% percentile inside its fair comparison set

0%Raw benchmark valueCI 0% - 0%

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 #7 · Source label: google/gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 78.8%
Percentile: 90.7%
Last updated: aging
Eligibility: preview_model

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

90.7% percentile inside its fair comparison set

78.8%Raw benchmark valueCI 75.2% - 82.4%

Terminal-Bench 2.1

VALS-AI · Coding · Objective

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

Rank #5 · Source label: google/gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 70.8%
Percentile: 85.2%
Last updated: aging
Eligibility: preview_model

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

85.2% percentile inside its fair comparison set

70.8%Raw benchmark valueCI 68.6% - 73%

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 #21 · Source label: google/gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 32%
Percentile: 59.2%
Last updated: aging
Eligibility: preview_model

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

59.2% percentile inside its fair comparison set

32%Raw benchmark valueCI 23.5% - 40.5%

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 #11 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,523
Percentile: 96.9%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: coding. Source rank: #14. Votes: 16359. Organization: google. License: Proprietary.

96.9% percentile inside its fair comparison set

1,523Raw benchmark valueCI 1,517 - 1,529

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 #16 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,486
Percentile: 95.3%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: coding. Source rank: #20. Votes: 16359. Organization: google. License: Proprietary.

95.3% percentile inside its fair comparison set

1,486Raw benchmark valueCI 1,480 - 1,492

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 #7 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,490
Percentile: 64.7%
Last updated: stale
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: image_to_webdev. Source rank: #9. Votes: 3597. Organization: google. License: Proprietary.

64.7% percentile inside its fair comparison set

1,490Raw benchmark valueCI 1,478 - 1,502

Debugging

BB · Coding · Rubric

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

Rank #7 · Source label: google/gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: BridgeBench
Raw value: 85.9%
Percentile: 75%
Last updated: stale
Eligibility: preview_model

Parsed from the BridgeBench page for bridgebench-debugging.

75% percentile inside its fair comparison set

85.9%Raw benchmark value

Refactoring

BB · Coding · Rubric

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

Rank #3 · Source label: google/gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: BridgeBench
Raw value: 70%
Percentile: 90.5%
Last updated: stale
Eligibility: preview_model

Parsed from the BridgeBench page for bridgebench-refactoring.

90.5% percentile inside its fair comparison set

70%Raw benchmark value

Security

BB · Coding · Rubric

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

Rank #3 · Source label: google/gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: BridgeBench
Raw value: 85.2%
Percentile: 92%
Last updated: stale
Eligibility: preview_model

Parsed from the BridgeBench page for bridgebench-security.

92% percentile inside its fair comparison set

85.2%Raw benchmark value

UI bench

BB · Coding · Rubric

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

Rank #19 · Source label: google/gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: BridgeBench
Raw value: 52.9%
Percentile: 18.2%
Last updated: stale
Eligibility: preview_model

Parsed from the BridgeBench page for bridgebench-ui.

18.2% percentile inside its fair comparison set

52.9%Raw benchmark value

Speed throughput

BB · Coding · Speed / cost

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

Rank #10 · Source label: google/gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: BridgeBench
Raw value: 122.2 t/s
Percentile: 64%
Last updated: stale
Eligibility: preview_model

Parsed from the BridgeBench page for bridgebench-throughput.

64% percentile inside its fair comparison set

122.2 t/sRaw benchmark value

Speed TTFT

BB · Coding · Speed / cost

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

Rank #21 · Source label: google/gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: BridgeBench
Raw value: 7608.00ms
Percentile: 20%
Last updated: stale
Eligibility: preview_model

Parsed from the BridgeBench page for bridgebench-ttft.

20% percentile inside its fair comparison set

7608.00msRaw benchmark value

Speed cost

BB · Coding · Speed / cost

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

Rank #15 · Source label: google/gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: BridgeBench
Raw value: $0.1 $
Percentile: 48%
Last updated: stale
Eligibility: preview_model

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

48% percentile inside its fair comparison set

$0.1 $Raw benchmark value

Reasoning / math / science9 benchmarks85.6%

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 #8 · Source label: Gemini 3.1 Pro Preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 47%
Percentile: 98.4%
Last updated: recent
Eligibility: preview_model

Parsed from Artificial Analysis public leaderboard field `hle`.

98.4% percentile inside its fair comparison set

47%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 #3 · Source label: Gemini 3.1 Pro Preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 94.1%
Percentile: 99.6%
Last updated: recent
Eligibility: preview_model

Parsed from Artificial Analysis public leaderboard field `gpqa`.

99.6% percentile inside its fair comparison set

94.1%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 #16 · Source label: Gemini 3.1 Pro Preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 17.7%
Percentile: 96.2%
Last updated: recent
Eligibility: preview_model

Parsed from Artificial Analysis public leaderboard field `critpt`.

96.2% percentile inside its fair comparison set

17.7%Raw benchmark value

ProofBench

VALS-AI · Reasoning / math / science · Objective

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

Rank #12 · Source label: google/gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 26%
Percentile: 71.4%
Last updated: aging
Eligibility: preview_model

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

71.4% percentile inside its fair comparison set

26%Raw benchmark valueCI 17.4% - 34.6%

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 #1 · Source label: google/gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 95.5%
Percentile: 100%
Last updated: aging
Eligibility: preview_model

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

100% percentile inside its fair comparison set

95.5%Raw benchmark valueCI 93.4% - 97.5%

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 #2 · Source label: google/gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 91%
Percentile: 98.9%
Last updated: aging
Eligibility: preview_model

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

98.9% percentile inside its fair comparison set

91%Raw benchmark valueCI 90.4% - 91.5%

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 #8 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,495
Percentile: 97.8%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: math. Source rank: #10. Votes: 3429. Organization: google. License: Proprietary.

97.8% percentile inside its fair comparison set

1,495Raw benchmark valueCI 1,484 - 1,506

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 #6 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,491
Percentile: 98.4%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: math. Source rank: #9. Votes: 3429. Organization: google. License: Proprietary.

98.4% percentile inside its fair comparison set

1,491Raw benchmark valueCI 1,481 - 1,502

Reasoning

BB · Reasoning / math / science · Rubric

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

Rank #20 · Source label: google/gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: BridgeBench
Raw value: 34.3%
Percentile: 9.5%
Last updated: stale
Eligibility: preview_model

Parsed from the BridgeBench page for bridgebench-reasoning.

9.5% percentile inside its fair comparison set

34.3%Raw benchmark value

Professional reasoning32 benchmarks79.9%

APEX-Agents-AA

AA · Professional reasoning · Objective

Long-horizon agentic task completion.

Rank #9 · Source label: Gemini 3.1 Pro Preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 32%
Percentile: 71.4%
Last updated: recent
Eligibility: preview_model

Parsed from Artificial Analysis public leaderboard field `apexAgents`.

71.4% percentile inside its fair comparison set

32%Raw benchmark value

GDPval-AA

AA · Professional reasoning · Rubric

Agentic performance on economically valuable work tasks.

Rank #79 · Source label: Gemini 3.1 Pro Preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 965
Percentile: 54.1%
Last updated: recent
Eligibility: preview_model

Parsed from Artificial Analysis public leaderboard field `gdpvalBreakdown.elo`.

54.1% percentile inside its fair comparison set

965Raw benchmark value

Legal Research Bench

VALS-AI · Professional reasoning · Objective

Applied legal research tasks.

Rank #10 · Source label: google/gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 20.7%
Percentile: 25%
Last updated: aging
Eligibility: preview_model

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

25% percentile inside its fair comparison set

20.7%Raw benchmark valueCI 15.2% - 26.2%

Public Benefits Bench

VALS-AI · Professional reasoning · Objective

Answering SNAP benefits questions across the public-benefits lifecycle.

Rank #10 · Source label: google/gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 53.8%
Percentile: 18.2%
Last updated: aging
Eligibility: preview_model

Parsed from Vals AI BenchmarkView overall scores. Vals slug: public-benefits-bench; provider: Google.

18.2% percentile inside its fair comparison set

53.8%Raw benchmark valueCI 51.2% - 56.3%

Vals Index

VALS-AI · Professional reasoning · Combined

Weighted model performance across economically relevant Vals tasks.

Rank #12 · Source label: google/gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 54
Percentile: 57.7%
Last updated: aging
Eligibility: preview_model

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

57.7% percentile inside its fair comparison set

54Raw benchmark valueCI 51 - 57

LegalBench

VALS-AI · Professional reasoning · Objective

Academic legal reasoning tasks.

Rank #2 · Source label: google/gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 87.4%
Percentile: 98.9%
Last updated: aging
Eligibility: preview_model

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

98.9% percentile inside its fair comparison set

87.4%Raw benchmark valueCI 86.8% - 88%

Finance Agent v2

VALS-AI · Professional reasoning · Objective

Core financial analyst tasks for agentic models.

Rank #13 · Source label: google/gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 43%
Percentile: 52%
Last updated: aging
Eligibility: preview_model

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

52% percentile inside its fair comparison set

43%Raw benchmark valueCI 40.6% - 45.3%

TaxEval v2

VALS-AI · Professional reasoning · Objective

Answer quality on tax questions and responses.

Rank #39 · Source label: google/gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 72.9%
Percentile: 58.2%
Last updated: aging
Eligibility: preview_model

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

58.2% 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 #1 · Source label: google/gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 59.1%
Percentile: 100%
Last updated: aging
Eligibility: preview_model

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

100% percentile inside its fair comparison set

59.1%Raw benchmark valueCI 55.1% - 63%

MedScribe

VALS-AI · Professional reasoning · Objective

Administrative documentation support for doctors.

Rank #31 · Source label: google/gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 76.1%
Percentile: 40%
Last updated: aging
Eligibility: preview_model

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

40% percentile inside its fair comparison set

76.1%Raw benchmark valueCI 72.4% - 79.9%

Text Arena · Expert

AR · Professional reasoning · Human

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

Rank #8 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,515
Percentile: 97.5%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: expert. Source rank: #10. Votes: 5352. Organization: google. License: Proprietary.

97.5% percentile inside its fair comparison set

1,515Raw benchmark valueCI 1,506 - 1,524

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 #11 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,477
Percentile: 96.9%
Last updated: aging
Eligibility: preview_model

96.9% percentile inside its fair comparison set

1,477Raw benchmark valueCI 1,471 - 1,484

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 #5 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,467
Percentile: 98.8%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: industry_entertainment_and_sports_and_media. Source rank: #7. Votes: 12538. Organization: google. License: Proprietary.

98.8% percentile inside its fair comparison set

1,467Raw benchmark valueCI 1,461 - 1,474

Text Arena · Industry Legal And Government

AR · Professional reasoning · Human

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

Rank #8 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,495
Percentile: 97.7%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: industry_legal_and_government. Source rank: #9. Votes: 4683. Organization: google. License: Proprietary.

97.7% percentile inside its fair comparison set

1,495Raw benchmark valueCI 1,486 - 1,505

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 #6 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,508
Percentile: 98.5%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: industry_life_and_physical_and_social_science. Source rank: #8. Votes: 9896. Organization: google. License: Proprietary.

98.5% percentile inside its fair comparison set

1,508Raw benchmark valueCI 1,501 - 1,515

Text Arena · Industry Mathematical

AR · Professional reasoning · Human

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

Rank #12 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,490
Percentile: 96.4%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: industry_mathematical. Source rank: #15. Votes: 3217. Organization: google. License: Proprietary.

96.4% percentile inside its fair comparison set

1,490Raw benchmark valueCI 1,479 - 1,501

Text Arena · Industry Medicine And Healthcare

AR · Professional reasoning · Human

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

Rank #6 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,505
Percentile: 98.3%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: industry_medicine_and_healthcare. Source rank: #8. Votes: 4474. Organization: google. License: Proprietary.

98.3% percentile inside its fair comparison set

1,505Raw benchmark valueCI 1,495 - 1,514

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 #9 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,514
Percentile: 97.5%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: industry_software_and_it_services. Source rank: #12. Votes: 23570. Organization: google. License: Proprietary.

97.5% percentile inside its fair comparison set

1,514Raw benchmark valueCI 1,509 - 1,519

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 #5 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,479
Percentile: 98.8%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: industry_writing_and_literature_and_language. Source rank: #7. Votes: 14515. Organization: google. License: Proprietary.

98.8% percentile inside its fair comparison set

1,479Raw benchmark valueCI 1,473 - 1,485

Text Arena · Expert · No Style Control

AR · Professional reasoning · Human

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

Rank #12 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,492
Percentile: 96%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: expert. Source rank: #15. Votes: 5352. Organization: google. License: Proprietary.

96% percentile inside its fair comparison set

1,492Raw benchmark valueCI 1,483 - 1,501

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 #12 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,461
Percentile: 96.5%
Last updated: aging
Eligibility: preview_model

96.5% percentile inside its fair comparison set

1,461Raw benchmark valueCI 1,454 - 1,467

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 #4 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,465
Percentile: 99.1%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: industry_entertainment_and_sports_and_media. Source rank: #6. Votes: 12538. Organization: google. License: Proprietary.

99.1% percentile inside its fair comparison set

1,465Raw benchmark valueCI 1,458 - 1,471

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 #6 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,485
Percentile: 98.3%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: industry_legal_and_government. Source rank: #7. Votes: 4683. Organization: google. License: Proprietary.

98.3% percentile inside its fair comparison set

1,485Raw benchmark valueCI 1,476 - 1,495

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 #6 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,497
Percentile: 98.5%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: industry_life_and_physical_and_social_science. Source rank: #7. Votes: 9896. Organization: google. License: Proprietary.

98.5% percentile inside its fair comparison set

1,497Raw benchmark valueCI 1,490 - 1,503

Text Arena · Industry Mathematical · No Style Control

AR · Professional reasoning · Human

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

Rank #14 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,481
Percentile: 95.8%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: industry_mathematical. Source rank: #16. Votes: 3217. Organization: google. License: Proprietary.

95.8% percentile inside its fair comparison set

1,481Raw benchmark valueCI 1,470 - 1,493

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 #6 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,488
Percentile: 98.3%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: industry_medicine_and_healthcare. Source rank: #8. Votes: 4474. Organization: google. License: Proprietary.

98.3% percentile inside its fair comparison set

1,488Raw benchmark valueCI 1,478 - 1,498

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 #13 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,488
Percentile: 96.3%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: industry_software_and_it_services. Source rank: #16. Votes: 23570. Organization: google. License: Proprietary.

96.3% percentile inside its fair comparison set

1,488Raw benchmark valueCI 1,483 - 1,493

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 #4 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,479
Percentile: 99.1%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: industry_writing_and_literature_and_language. Source rank: #5. Votes: 14515. Organization: google. License: Proprietary.

99.1% percentile inside its fair comparison set

1,479Raw benchmark valueCI 1,473 - 1,486

SAGE

VALS-AI · Professional reasoning · Objective

Student Assessment with Generative Evaluation.

Rank #16 · Source label: google/gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 48.7%
Percentile: 66.7%
Last updated: aging
Eligibility: preview_model

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

66.7% percentile inside its fair comparison set

48.7%Raw benchmark valueCI 42.2% - 55.1%

Public Benefits Bench v1

VALS-AI · Professional reasoning · Objective

Answering public-benefits questions across the benefits lifecycle.

Rank #10 · Source label: google/gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 50.9%
Percentile: 25%
Last updated: aging
Eligibility: preview_model

Parsed from Vals AI BenchmarkView overall scores. Vals slug: public-benefits-bench-v1; provider: Google.

25% percentile inside its fair comparison set

50.9%Raw benchmark valueCI 48.3% - 53.4%

Hallucination

BB · Professional reasoning · Rubric

Accuracy and fabrication resistance under prompts that invite unsupported claims.

Rank #2 · Source label: google/gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: BridgeBench
Raw value: 79.1%
Percentile: 97.1%
Last updated: stale
Eligibility: preview_model

Parsed from the BridgeBench page for bridgebench-hallucination.

97.1% percentile inside its fair comparison set

79.1%Raw benchmark value

BS pushback

BB · Professional reasoning · Rubric

Resistance to confidently accepting bogus assumptions in expert-style prompts.

Rank #13 · Source label: google/gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: BridgeBench
Raw value: 66.5%
Percentile: 33.3%
Last updated: stale
Eligibility: preview_model

Parsed from the BridgeBench page for bridgebench-pushback.

33.3% percentile inside its fair comparison set

66.5%Raw benchmark value

Search / tool use1 benchmark96.4%

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 #13 · Source label: Gemini 3.1 Pro Preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 95.6%
Percentile: 96.4%
Last updated: recent
Eligibility: preview_model

Parsed from Artificial Analysis public leaderboard field `tau2`.

96.4% percentile inside its fair comparison set

95.6%Raw benchmark value

Long context2 benchmarks85.5%

Long Context Reasoning

AA · Long context · Objective

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

Rank #9 · Source label: Gemini 3.1 Pro Preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 79%
Percentile: 98.2%
Last updated: recent
Eligibility: preview_model

Parsed from Artificial Analysis public leaderboard field `lcr`.

98.2% percentile inside its fair comparison set

79%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 #25 · Source label: google/gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 64.5%
Percentile: 72.7%
Last updated: aging
Eligibility: preview_model

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

72.7% percentile inside its fair comparison set

64.5%Raw benchmark valueCI 62.6% - 66.3%

Vision understanding20 benchmarks92.2%

MMMU-Pro

AA · Vision understanding · Objective

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

Rank #8 · Source label: Gemini 3.1 Pro Preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 82.4%
Percentile: 96.3%
Last updated: recent
Eligibility: preview_model

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

96.3% percentile inside its fair comparison set

82.4%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 #10 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,280
Percentile: 91.7%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: overall. Source rank: #13. Votes: 26478. Organization: google. License: Proprietary.

91.7% percentile inside its fair comparison set

1,280Raw benchmark valueCI 1,273 - 1,286

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 #3 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,258
Percentile: 92.3%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: captioning. Source rank: #3. Votes: 97. Organization: google. License: Proprietary.

92.3% percentile inside its fair comparison set

1,258Raw benchmark valueCI 1,205 - 1,311

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 #5 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,294
Percentile: 92.7%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: creative_writing_vision. Source rank: #6. Votes: 1507. Organization: google. License: Proprietary.

92.7% percentile inside its fair comparison set

1,294Raw benchmark valueCI 1,278 - 1,311

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 #8 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,311
Percentile: 90%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: diagram. Source rank: #11. Votes: 6941. Organization: google. License: Proprietary.

90% percentile inside its fair comparison set

1,311Raw benchmark valueCI 1,302 - 1,320

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 #14 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,272
Percentile: 88.1%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: english. Source rank: #17. Votes: 11106. Organization: google. License: Proprietary.

88.1% percentile inside its fair comparison set

1,272Raw benchmark valueCI 1,263 - 1,281

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 #3 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,293
Percentile: 93.8%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: entity_recognition. Source rank: #3. Votes: 218. Organization: google. License: Proprietary.

93.8% percentile inside its fair comparison set

1,293Raw benchmark valueCI 1,255 - 1,331

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 #8 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,311
Percentile: 89.7%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: homework. Source rank: #11. Votes: 4011. Organization: google. License: Proprietary.

89.7% percentile inside its fair comparison set

1,311Raw benchmark valueCI 1,301 - 1,322

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 #5 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,300
Percentile: 91.8%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: humor. Source rank: #5. Votes: 994. Organization: google. License: Proprietary.

91.8% percentile inside its fair comparison set

1,300Raw benchmark valueCI 1,280 - 1,319

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 #9 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,297
Percentile: 88.6%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: ocr. Source rank: #11. Votes: 18511. Organization: google. License: Proprietary.

88.6% percentile inside its fair comparison set

1,297Raw benchmark valueCI 1,291 - 1,303

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 #7 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,297
Percentile: 94.5%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: overall. Source rank: #9. Votes: 26478. Organization: google. License: Proprietary.

94.5% percentile inside its fair comparison set

1,297Raw benchmark valueCI 1,291 - 1,303

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 #2 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,316
Percentile: 96.2%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: captioning. Source rank: #2. Votes: 97. Organization: google. License: Proprietary.

96.2% percentile inside its fair comparison set

1,316Raw benchmark valueCI 1,263 - 1,369

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 #5 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,317
Percentile: 92.7%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: creative_writing_vision. Source rank: #6. Votes: 1507. Organization: google. License: Proprietary.

92.7% percentile inside its fair comparison set

1,317Raw benchmark valueCI 1,301 - 1,333

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 #6 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,315
Percentile: 92.9%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: diagram. Source rank: #9. Votes: 6941. Organization: google. License: Proprietary.

92.9% percentile inside its fair comparison set

1,315Raw benchmark valueCI 1,306 - 1,324

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 #12 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,288
Percentile: 89.9%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: english. Source rank: #14. Votes: 11106. Organization: google. License: Proprietary.

89.9% percentile inside its fair comparison set

1,288Raw benchmark valueCI 1,279 - 1,296

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 #3 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,335
Percentile: 93.8%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: entity_recognition. Source rank: #3. Votes: 218. Organization: google. License: Proprietary.

93.8% percentile inside its fair comparison set

1,335Raw benchmark valueCI 1,297 - 1,373

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 #7 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,318
Percentile: 91.2%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: homework. Source rank: #10. Votes: 4011. Organization: google. License: Proprietary.

91.2% percentile inside its fair comparison set

1,318Raw benchmark valueCI 1,308 - 1,328

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 #4 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,323
Percentile: 93.9%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: humor. Source rank: #4. Votes: 994. Organization: google. License: Proprietary.

93.9% percentile inside its fair comparison set

1,323Raw benchmark valueCI 1,303 - 1,343

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 #8 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,308
Percentile: 90%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: ocr. Source rank: #10. Votes: 18511. Organization: google. License: Proprietary.

90% percentile inside its fair comparison set

1,308Raw benchmark valueCI 1,302 - 1,314

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 #4 · Source label: google/gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 88.2%
Percentile: 94.8%
Last updated: aging
Eligibility: preview_model

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

94.8% percentile inside its fair comparison set

88.2%Raw benchmark valueCI 86.7% - 89.7%

Document understanding4 benchmarks64.7%

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 #12 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,455
Percentile: 54.2%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: overall. Source rank: #15. Votes: 38090. Organization: google. License: Proprietary.

54.2% percentile inside its fair comparison set

1,455Raw benchmark valueCI 1,450 - 1,461

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 #13 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,441
Percentile: 50%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: overall. Source rank: #16. Votes: 38090. Organization: google. License: Proprietary.

50% percentile inside its fair comparison set

1,441Raw benchmark valueCI 1,436 - 1,447

Vals Multimodal Index

VALS-AI · Document understanding · Combined

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

Rank #9 · Source label: google/gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 56
Percentile: 57.9%
Last updated: aging
Eligibility: preview_model

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

57.9% percentile inside its fair comparison set

56Raw benchmark valueCI 53 - 59

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 #3 · Source label: google/gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 69.4%
Percentile: 96.7%
Last updated: aging
Eligibility: preview_model

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

96.7% percentile inside its fair comparison set

69.4%Raw benchmark valueCI 67.6% - 71.2%

Multilingual16 benchmarks98.1%

Text Arena · Chinese

AR · Multilingual · Human

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

Rank #7 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,528
Percentile: 98%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: chinese. Source rank: #9. Votes: 3203. Organization: google. License: Proprietary.

98% percentile inside its fair comparison set

1,528Raw benchmark valueCI 1,516 - 1,539

Text Arena · French

AR · Multilingual · Human

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

Rank #9 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,505
Percentile: 96.3%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: french. Source rank: #10. Votes: 1967. Organization: google. License: Proprietary.

96.3% percentile inside its fair comparison set

1,505Raw benchmark valueCI 1,489 - 1,521

Text Arena · German

AR · Multilingual · Human

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

Rank #5 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,497
Percentile: 98.3%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: german. Source rank: #5. Votes: 1012. Organization: google. License: Proprietary.

98.3% percentile inside its fair comparison set

1,497Raw benchmark valueCI 1,477 - 1,517

Text Arena · Japanese

AR · Multilingual · Human

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

Rank #2 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,510
Percentile: 99.5%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: japanese. Source rank: #2. Votes: 535. Organization: google. License: Proprietary.

99.5% percentile inside its fair comparison set

1,510Raw benchmark valueCI 1,482 - 1,538

Text Arena · Korean

AR · Multilingual · Human

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

Rank #4 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,466
Percentile: 98.6%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: korean. Source rank: #4. Votes: 1054. Organization: google. License: Proprietary.

98.6% percentile inside its fair comparison set

1,466Raw benchmark valueCI 1,445 - 1,487

Text Arena · Russian

AR · Multilingual · Human

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

Rank #6 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,501
Percentile: 98.3%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: russian. Source rank: #7. Votes: 6426. Organization: google. License: Proprietary.

98.3% percentile inside its fair comparison set

1,501Raw benchmark valueCI 1,492 - 1,509

Text Arena · Spanish

AR · Multilingual · Human

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

Rank #3 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,484
Percentile: 99.1%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: spanish. Source rank: #4. Votes: 1960. Organization: google. License: Proprietary.

99.1% percentile inside its fair comparison set

1,484Raw benchmark valueCI 1,468 - 1,499

Text Arena · Chinese · No Style Control

AR · Multilingual · Human

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

Rank #7 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,528
Percentile: 98%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: chinese. Source rank: #10. Votes: 3203. Organization: google. License: Proprietary.

98% percentile inside its fair comparison set

1,528Raw benchmark valueCI 1,516 - 1,539

Text Arena · French · No Style Control

AR · Multilingual · Human

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

Rank #12 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,490
Percentile: 94.9%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: french. Source rank: #14. Votes: 1967. Organization: google. License: Proprietary.

94.9% percentile inside its fair comparison set

1,490Raw benchmark valueCI 1,475 - 1,506

Text Arena · German · No Style Control

AR · Multilingual · Human

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

Rank #4 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,496
Percentile: 98.7%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: german. Source rank: #4. Votes: 1012. Organization: google. License: Proprietary.

98.7% percentile inside its fair comparison set

1,496Raw benchmark valueCI 1,476 - 1,515

Text Arena · Japanese · No Style Control

AR · Multilingual · Human

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

Rank #4 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,507
Percentile: 98.5%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: japanese. Source rank: #4. Votes: 535. Organization: google. License: Proprietary.

98.5% percentile inside its fair comparison set

1,507Raw benchmark valueCI 1,479 - 1,535

Text Arena · Korean · No Style Control

AR · Multilingual · Human

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

Rank #3 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,468
Percentile: 99%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: korean. Source rank: #3. Votes: 1054. Organization: google. License: Proprietary.

99% percentile inside its fair comparison set

1,468Raw benchmark valueCI 1,447 - 1,488

Text Arena · Russian · No Style Control

AR · Multilingual · Human

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

Rank #3 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,497
Percentile: 99.3%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: russian. Source rank: #3. Votes: 6426. Organization: google. License: Proprietary.

99.3% percentile inside its fair comparison set

1,497Raw benchmark valueCI 1,489 - 1,505

Text Arena · Spanish · No Style Control

AR · Multilingual · Human

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

Rank #2 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,483
Percentile: 99.5%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: spanish. Source rank: #3. Votes: 1960. Organization: google. License: Proprietary.

99.5% percentile inside its fair comparison set

1,483Raw benchmark valueCI 1,468 - 1,498

Vision Arena · Chinese

AR · Multilingual · Human

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

Rank #4 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,344
Percentile: 96.1%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: chinese. Source rank: #5. Votes: 1599. Organization: google. License: Proprietary.

96.1% percentile inside its fair comparison set

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

Vision Arena · Chinese · No Style Control

AR · Multilingual · Human

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

Rank #3 · Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,369
Percentile: 97.4%
Last updated: aging
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `gemini-3.1-pro-preview`. Category: chinese. Source rank: #3. Votes: 1599. Organization: google. License: Proprietary.

97.4% percentile inside its fair comparison set

1,369Raw benchmark valueCI 1,349 - 1,389