supermarket

Vision-language

Gemini 3.5 Flash

By Google DeepMind, in the Gemini Flash family.

Released
19 May 2026
Status
Legacy
Context
1M
Max output
65.5K
Input per 1M tokens
$1.50
Output per 1M tokens
$9.00
Open weights
No
License
proprietary
Parameters
—
Knowledge cutoff
Jan 2025
Reasoning
Yes
Modalities
text, image, video, audio, pdf → text

Category scores

Out of 100, from independent and third-party results only, by method category-2026-09-29.1.

Category scores
CategoryScoreRankCoverage
Knowledge & reasoning44.9284 of 4
Coding & software engineering37.4Not ranked1 of 5
Factuality & hallucination66.2101 of 1
Human preference54.3174 of 5

Benchmark results

Every result we hold for this model, with who ran it and where it was published.

Knowledge & reasoning

Knowledge & reasoning results
BenchmarkMetricValueProvenanceSettingEvaluatedSource
Chess PuzzlesPuzzles solved50.0%Independenthigh28 May 2026Epoch AI
Chess PuzzlesPuzzles solved45.0%Independentlow6 Aug 2026Epoch AI
Chess PuzzlesPuzzles solved43.0%Independentminimal15 Jul 2026Epoch AI
DTBench (Decision Theory Benchmark)Accuracy (capability questions)94.7%Third partyhigh—https://conceptualreasoning.ai/dtbench via Epoch AI
Earthborne Rangers (EBR-bench)Objectives reached4.8%Independenthigh25 Jun 2026Epoch AI
LMCA (Language Model Conceptual Argumentation)Correlation with expert ratings (Pearson r)0.4715Third partyhigh—https://conceptualreasoning.ai/lmca via Epoch AI
Mystery Game PuzzlesPuzzles solved32.0%Independenthigh27 Jul 2026Epoch AI
Mystery Game PuzzlesPuzzles solved28.0%Independentlow5 Aug 2026Epoch AI
SimpleBenchScore (AVG@5)76.7%Third partyunknown—SimpleBench Leaderboard via Epoch AI

Scientific reasoning

Scientific reasoning results
BenchmarkMetricValueProvenanceSettingEvaluatedSource
GPQA DiamondAccuracy92.8%Independenthigh22 May 2026Epoch AI
GPQA DiamondAccuracy88.9%Independentlow6 Aug 2026Epoch AI
GPQA DiamondAccuracy86.4%Independentminimal15 Jul 2026Epoch AI
Surface Evolver BenchMean score (partial credit)58.1%Third partymedium—https://yhenon.github.io/surface-evolver-llm-eval/ via Epoch AI

Mathematics

Mathematics results
BenchmarkMetricValueProvenanceSettingEvaluatedSource
FrontierMath Tier 4Accuracy26.8%Independenthigh10 Jun 2026Epoch AI
FrontierMath Tier 4 (v1)Accuracy14.6%Independenthigh25 May 2026Epoch AI
FrontierMath Tiers 1-3Accuracy62.8%Independenthigh10 Jun 2026Epoch AI
FrontierMath Tiers 1-3 (v1)Accuracy39.0%Independenthigh22 May 2026Epoch AI
OTIS Mock AIME 2024-2025Accuracy95.6%Independenthigh25 May 2026Epoch AI
OTIS Mock AIME 2024-2025Accuracy88.9%Independentlow6 Aug 2026Epoch AI
OTIS Mock AIME 2024-2025Accuracy80.0%Independentminimal15 Jul 2026Epoch AI

Coding & software engineering

Coding & software engineering results
BenchmarkMetricValueProvenanceSettingEvaluatedSource
DeepSWEPass@137.4%Third partymedium—https://deepswe.datacurve.ai/ via Epoch AI
DeepSWEPass@136.1%Third partyhigh—https://deepswe.datacurve.ai/ via Epoch AI
SWE-bench Verified% resolved79.3%Independenthigh1 Jun 2026Epoch AI
WeirdML (v2)Average accuracy62.6%Third partyhigh—https://htihle.github.io/weirdml.html via Epoch AI

Professional knowledge work

Professional knowledge work results
BenchmarkMetricValueProvenanceSettingEvaluatedSource
APEX-AgentsPass@1 (all rubric criteria met)27.5%Source not statedunknown—Epoch AI via Epoch AI

Factuality & hallucination

Factuality & hallucination results
BenchmarkMetricValueProvenanceSettingEvaluatedSource
SimpleQA VerifiedAccuracy (% correct)66.2%Independenthigh27 Aug 2026Epoch AI

Human preference

Human preference results
BenchmarkMetricValueProvenanceSettingEvaluatedSource
Code Arena: WebDev (arena.ai)Arena score (Elo scale)1499Independenthigh25 Sep 2026Arena
Code Arena: WebDev (arena.ai)Arena score (Elo scale)1490Independentmedium25 Sep 2026Arena
Code Arena: WebDev (arena.ai)Votes12270Independentmedium25 Sep 2026Arena
Code Arena: WebDev (arena.ai)Votes9530Independenthigh25 Sep 2026Arena
Document Arena (arena.ai)Arena score (Elo scale)1461Independentmedium13 Sep 2026Arena
Document Arena (arena.ai)Arena score (Elo scale)1453Independenthigh13 Sep 2026Arena
Document Arena (arena.ai)Votes6500Independentmedium13 Sep 2026Arena
Document Arena (arena.ai)Votes4061Independenthigh13 Sep 2026Arena
Text Arena (arena.ai)Arena score (Elo scale)1477Independenthigh25 Sep 2026Arena
Text Arena (arena.ai)Arena score (Elo scale)1474Independentmedium25 Sep 2026Arena
Text Arena (arena.ai)Votes46190Independenthigh25 Sep 2026Arena
Text Arena (arena.ai)Votes44906Independentmedium25 Sep 2026Arena
Vision Arena (arena.ai)Arena score (Elo scale)1285Independenthigh28 Sep 2026Arena
Vision Arena (arena.ai)Arena score (Elo scale)1284Independentmedium28 Sep 2026Arena
Vision Arena (arena.ai)Votes12124Independentmedium28 Sep 2026Arena
Vision Arena (arena.ai)Votes11983Independenthigh28 Sep 2026Arena

Other Gemini Flash versions

Sources