APEX-Agents-AA ↗47.05%
Gemini 3.5 Flash (High) · APEX-Agents-AA ↗
Google / MODEL EVIDENCE
Gemini 3.5 Flash
64 benchmarks · 83 observations · 69 sources · released 2026-05-19
Scores at a glance
64 benchmarks · large dot: Gemini 3.5 Flash; small dots: closest models where reported
AA-AnalystAgent ↗45.00%
Gemini 3.5 Flash (High) · AA-AnalystAgent · strict score ↗
EnterpriseOps-Gym-AA ↗50.07%
Gemini 3.5 Flash (High) · EnterpriseOps-Gym-AA ↗
ITBench-AA ↗40.35%
Gemini 3.5 Flash (High) · ITBench-AA · root-cause score ↗
MMMU-Pro ↗88.27 %
Gemini 3.5 Flash · MMMU Pro · Accuracy ↗
1 more reported result
84.28% · Gemini 3.5 Flash (High) · MMMU-Pro · AA implementation · Artificial Analysis ↗
IFBench ↗76.33%
Gemini 3.5 Flash (High) · IFBench · AA implementation ↗
τ²-Bench Telecom ↗95.61%
Gemini 3.5 Flash (Medium) · τ²-Bench Telecom ↗
GPQA Diamond ↗92.8%
Gemini 3.5 Flash · GPQA diamond · Best score (across scorers) · % ↗
3 more reported results
88.89% · Gemini 3.5 Flash · GPQA diamond · Best score (across scorers) · % · Epoch AI · GPQA diamond ↗
86.36% · Gemini 3.5 Flash · GPQA diamond · Best score (across scorers) · % · Epoch AI · GPQA diamond ↗
92.68 % · Gemini 3.5 Flash · GPQA Diamond · Accuracy · Vals AI · GPQA Diamond ↗
Gemini 3.5 Flash · FrontierMath-2025-02-28-Private · Best score (across scorers) · % ↗
Gemini 3.5 Flash · FrontierMath-Tier-4-2025-07-01-Private · Best score (across scorers) · % ↗
Gemini 3.5 Flash · FrontierMath-Tiers-1-3-v2-Private · Best score (across scorers) · % ↗
Gemini 3.5 Flash · FrontierMath-Tier-4-v2-Private · Best score (across scorers) · % ↗
Gemini 3.5 Flash · OTIS Mock AIME 2024-2025 · Best score (across scorers) · % ↗
2 more reported results
88.89% · Gemini 3.5 Flash · OTIS Mock AIME 2024-2025 · Best score (across scorers) · % · Epoch AI · OTIS Mock AIME 2024-2025 ↗
80% · Gemini 3.5 Flash · OTIS Mock AIME 2024-2025 · Best score (across scorers) · % · Epoch AI · OTIS Mock AIME 2024-2025 ↗
SWE-bench Verified ↗79.34%
Gemini 3.5 Flash · SWE-Bench verified · Best score (across scorers) · % ↗
1 more reported result
78.8 % · Gemini 3.5 Flash (Mini-SWE-agent) · SWE-bench Verified · Score · Vals AI · SWE-bench Verified ↗
SimpleQA Verified ↗66.2%
Gemini 3.5 Flash · SimpleQA Verified · Best score (across scorers) · % ↗
Gemini 3.5 Flash · Chess Puzzles · Best score (across scorers) · % ↗
2 more reported results
42% · Gemini 3.5 Flash · Chess Puzzles · Best score (across scorers) · % · Epoch AI · Chess Puzzles ↗
27% · Gemini 3.5 Flash · Chess Puzzles · Best score (across scorers) · % · Epoch AI · Chess Puzzles ↗
Gemini 3.5 Flash · EBR-bench · Best score (across scorers) · % ↗
Gemini 3.5 Flash · Mystery Game Puzzles · Best score (across scorers) · % ↗
1 more reported result
28% · Gemini 3.5 Flash · Mystery Game Puzzles · Best score (across scorers) · % · Epoch AI · Mystery Game Puzzles ↗
ARC-AGI-1 ↗92.5%
Gemini 3.5 Flash · ARC-AGI · Score · % ↗
1 more reported result
48.83% · Gemini 3.5 Flash · ARC-AGI · Score · % · Epoch AI · ARC-AGI ↗
SimpleBench ↗76.7%
Gemini 3.5 Flash · SimpleBench · Score (AVG@5) · % ↗
WeirdML (v2) ↗62.64%
Gemini 3.5 Flash · WeirdML · Accuracy · % ↗
Gemini 3.5 Flash · APEX-Agents · Pass@1 score · % ↗
ARC-AGI-2 ↗72.08%
Gemini 3.5 Flash · ARC-AGI-2 · Score · % ↗
1 more reported result
8.89% · Gemini 3.5 Flash · ARC-AGI-2 · Score · % · Epoch AI · ARC-AGI-2 ↗
GBAEval ↗6.7%
Gemini 3.5 Flash · GBAEval · Overall score · % ↗
CritPt ↗13.14%
Gemini 3.5 Flash · CritPt · Accuracy · % ↗
Surface Evolver Bench ↗58.13%
Gemini 3.5 Flash · Surface Evolver Bench · Mean score · % ↗
DeepSWE v1.1 ↗37.39%
Gemini 3.5 Flash · DeepSWE · Pass@1 · % ↗
1 more reported result
36.06% · Gemini 3.5 Flash · DeepSWE · Pass@1 · % · Epoch AI · DeepSWE ↗
ProofBench ↗31 %
Gemini 3.5 Flash · ProofBench · Accuracy ↗
1 more reported result
31% · Gemini 3.5 Flash · ProofBench · Accuracy · % · Epoch AI · ProofBench ↗
LMCA ↗47.15%
Gemini 3.5 Flash · LMCA · Score · % ↗
DTBench ↗94.67%
Gemini 3.5 Flash · DTBench · Accuracy · % ↗
GDP.pdf ↗14%
Gemini 3.5 Flash · GDP.pdf · GDP.pdf score · % ↗
SciCode ↗53.94%
Gemini 3.5 Flash · SciCode · Score · % ↗
Text Arena (Coding) ↗1506.38 Elo
Gemini 3.5 Flash · WebDev Arena · Elo ↗
2 more reported results
1500.04 Elo · Gemini 3.5 Flash · WebDev Arena · Elo · Epoch AI · WebDev Arena ↗
1491.55 Elo · Gemini 3.5 Flash · WebDev Arena · Elo · Epoch AI · WebDev Arena ↗
Blueprint-Bench 2 ↗33.62%
Gemini 3.5 Flash · Blueprint-Bench 2 · Score · % ↗
EnigmaEval ↗25.41%
Gemini 3.5 Flash · EnigmaEval · Accuracy · % ↗
Epoch Capabilities Index (ECI) ↗154.88 ECI
Gemini 3.5 Flash · General ECI · 26 October 2026 fit ↗
Vals Multimodal Index ↗62.93 %
Gemini 3.5 Flash · Vals Multimodal Index · Accuracy ↗
Legal Research Bench ↗30.77 %
Gemini 3.5 Flash · Legal Research Bench · All-Pass ↗
LegalBench ↗83.6 %
Gemini 3.5 Flash · LegalBench · Accuracy ↗
Gemini 3.5 Flash · Harvey's Legal Agent Benchmark v1 · Accuracy ↗
Tax Agent Bench ↗21.65 %
Gemini 3.5 Flash · Tax Agent Bench · Accuracy ↗
Vals Finance Agent v2 ↗57.86 %
Gemini 3.5 Flash · Finance Agent v2 · Accuracy ↗
Gemini 3.5 Flash · Excel Modeling Benchmark v1 · Accuracy ↗
CorpFin v2 ↗64.69 %
Gemini 3.5 Flash · CorpFin · Accuracy ↗
MortgageTax & TaxEval v2 ↗68.12 %
Gemini 3.5 Flash · MortgageTax · Accuracy ↗
1 more reported result
74.37 % · Gemini 3.5 Flash · TaxEval · Accuracy · Vals AI · TaxEval ↗
MedCode ↗55.83 %
Gemini 3.5 Flash · MedCode · Accuracy ↗
MedScribe ↗76.57 %
Gemini 3.5 Flash · MedScribe · Accuracy ↗
Gemini 3.5 Flash (Mini-SWE-agent) · Terminal-Bench-Science · Score ↗
MMLU-Pro ↗89.52 %
Gemini 3.5 Flash · MMLU Pro · Accuracy ↗
Gemini 3.5 Flash · SAGE · Accuracy ↗
Vibe Code Bench ↗48.68 %
Gemini 3.5 Flash (OpenHands) · Vibe Code Bench · original task set ↗
Gemini 3.5 Flash · ProgramBench · Accuracy ↗
Terminal-Bench 4.0 ↗6.06 %
Gemini 3.5 Flash (Mini-SWE-agent) · Terminal-Bench 4.0 · Score ↗
SkillsBench ↗52.74 %
Gemini 3.5 Flash (OpenHands) · SkillsBench · Score ↗
Code Migration v1.1 ↗26.75 %
Gemini 3.5 Flash · Code Migration v1 · Accuracy ↗
LiveCodeBench ↗87.6 %
Gemini 3.5 Flash · LiveCodeBench · Accuracy ↗
Public Benefits Bench v1.1 ↗59.47 %
Gemini 3.5 Flash · Public Benefits Bench · Accuracy ↗
1 more reported result
57.98 % · Gemini 3.5 Flash · Public Benefits Bench v1 · Accuracy · Vals AI · Public Benefits Bench v1 ↗
Gemini-3.5-Flash (high) · Humanity's Sixth Sense · Score ↗
DistressBench ↗66.89%
gemini-3.5-flash · DistressBench · Score ↗
DrugDiscoveryBench ↗48.80%
Gemini 3.5 Flash (mini-SWE-agent) high · DrugDiscoveryBench · Score ↗
Gemini 3.5 Flash · HiL-Bench (Human-in-Loop Benchmark) · Score ↗
MCP Atlas ↗83.60%
Gemini 3.5 Flash (high) · Scale Labs · Score ↗
ForecastBench ↗64.3 Brier index
gemini-3.5-flash† · Overall difficulty-adjusted Brier index ↗
Vending-Bench 2 ↗5396.41833 USD
Gemini 3.5 Flash · End-of-year balance ↗
Closest models by score
Similar positions on shared benchmark leaderboards. Smaller gaps are closer.
Kimi K2.6 →Kimi
25 of 41 shared benchmark comparisons
Claude Opus 4.7 →Anthropic
27 of 41 shared benchmark comparisons
Claude Sonnet 5 →Anthropic
26 of 41 shared benchmark comparisons
How similarity is measured
Mean absolute gap in percentile positions on shared benchmark settings; lower is closer. At least 60% overlap is required, with a minimum of three when available. Independent evaluations are used when there are at least three; otherwise published report comparisons are used. Repeated model observations within one setting are averaged before ranking. Writing-style traits and composite indices are excluded. This describes similarity, not an overall quality ranking.
Sources
Every plotted observation links to its evaluation. Reported configurations, benchmark versions and task subsets can differ; similarity describes shared scores, not training lift.
- Blueprint-Bench 2 | Andon Labs ↗
- ARC Prize - Leaderboard ↗
- Arena leaderboard — WebDev evaluation ↗
- AA-AnalystAgent Benchmark | Artificial Analysis ↗
- APEX-Agents-AA Benchmark | Artificial Analysis ↗
- EnterpriseOps-Gym-AA Benchmark | Artificial Analysis ↗
- IFBench Benchmark | Artificial Analysis ↗
- ITBench-AA Benchmark | Artificial Analysis ↗
- MMMU-Pro Benchmark | Artificial Analysis ↗
- SciCode Benchmark | Artificial Analysis ↗
- 𝜏²-Bench Telecom Benchmark | Artificial Analysis ↗
- DTBench — Conceptual Reasoning Index ↗
- LMCA — Conceptual Reasoning Index ↗
- DeepSWE ↗
- APEX-Agents ↗
- ARC-AGI-2 ↗
- Chess Puzzles ↗
- CritPt v1 ↗
- Earthborne Rangers (EBR-bench) ↗
- Forecastbench — methodology and published results ↗
- FrontierMath Tier 4 (v1, old) ↗
- FrontierMath Tier 4 (v2) ↗
- FrontierMath Tiers 1-3 (v1, old) ↗
- FrontierMath Tiers 1-3 (v2) ↗
- GPQA Diamond ↗
- Mystery Game Puzzles ↗
- OTIS Mock AIME 2024-2025 ↗
- SimpleQA Verified ↗
- SimpleBench ↗
- SWE-bench Verified ↗
- Vending Bench 2 — methodology and published results ↗
- AI benchmarks & capabilities ↗
- Gemini 3.5 Flash — GBA Eval ↗
- WeirdML ↗
- DistressBench — evaluation methodology and results ↗
- DrugDiscoveryBench — evaluation methodology and results ↗
- EnigmaEval ↗
- HiL-Bench (Human-in-Loop Benchmark) — evaluation methodology and results ↗
- Humanity's Sixth Sense — evaluation methodology and results ↗
- Scale Labs — evaluation methodology and results ↗
- GDP.pdf Benchmark | Surge AI ↗
- Code Migration v1 — evaluation methodology and results ↗
- CorpFin — evaluation methodology and results ↗
- Excel Modeling Benchmark v1 — evaluation methodology and results ↗
- Finance Agent v2 — evaluation methodology and results ↗
- GPQA Diamond — evaluation methodology and results ↗
- Harvey's Legal Agent Benchmark v1 — evaluation methodology and results ↗
- LiveCodeBench — evaluation methodology and results ↗
- LegalBench — evaluation methodology and results ↗
- Legal Research Bench — evaluation methodology and results ↗
- MedCode — evaluation methodology and results ↗
- MedScribe — evaluation methodology and results ↗
- MMLU Pro — evaluation methodology and results ↗
- MMMU Pro — evaluation methodology and results ↗
- MortgageTax — evaluation methodology and results ↗
- ProgramBench — evaluation methodology and results ↗
- ProofBench — evaluation methodology and results ↗
- Public Benefits Bench — evaluation methodology and results ↗
- Public Benefits Bench v1 — evaluation methodology and results ↗
- SAGE — evaluation methodology and results ↗
- SkillsBench — evaluation methodology and results ↗
- SWE-bench Verified — evaluation methodology and results ↗
- Tax Agent Bench — evaluation methodology and results ↗
- TaxEval — evaluation methodology and results ↗
- Terminal-Bench 4.0 — evaluation methodology and results ↗
- Terminal-Bench-Science — evaluation methodology and results ↗
- Vals Multimodal Index — evaluation methodology and results ↗
- Vibe Code Bench — evaluation methodology and results ↗
- Surface Evolver Bench ↗
Names used in source tables
Gemini 3.5 Flash · Gemini 3.5 Flash (High) · Gemini 3.5 Flash (Medium) · Gemini 3.5 Flash (Mini-SWE-agent) · Gemini 3.5 Flash (OpenHands) · Gemini 3.5 Flash (high) · Gemini 3.5 Flash (mini-SWE-agent) high · Gemini-3.5-Flash (high) · gemini-3.5-flash · gemini-3.5-flash†