Google · model
Gemini 3.5 Flash benchmark scores
As of 2026-11-22, Gemini 3.5 Flash (Google) has measured scores on 54 benchmarks and estimated scores on 141 more.
| Benchmark | Score | Source |
|---|---|---|
| AA-GPQA Diamond | 92.2% | measured |
| AA-HLE | 42.7% | measured |
| AA-SciCode | 53.9% | measured |
| AIME26 | 97.7% | estimated ± 1.2 pp, medium confidence |
| Artificial Analysis Intelligence Index | 50.2% | measured |
| CritPt | 13.1% | measured |
| GDPval-AA | 42.2% | measured |
| GPQA Diamond | 92.7% | measured |
| HLE | 40.2% | measured |
| IFBench | 76.3% | measured |
| AA-LCR | 69.3% | measured |
| LiveCodeBench v6 | 89.9% | estimated ± 2.2 pp, low confidence |
| AA-Omniscience Accuracy | 51.9% | measured |
| SciCode | 53.1% | measured |
| Terminal-Bench 2.1 | 76.2% | measured |
| BrowseComp | 90.2% | estimated ± 4.2 pp, medium confidence |
| HLE w/ tools | 57.0% | estimated ± 4.1 pp, low confidence |
| IFEval | 93.1% | estimated ± 2.8 pp, high confidence |
| LongBench v2 | 58.1% | estimated ± 5.5 pp, medium confidence |
| VITA-Bench | 49.6% | estimated ± 9.9 pp, low confidence |
| τ²-bench results | 95.3% | measured |
| AA Coding Index | 70.1% | measured |
| AA-MMMU-Pro | 84.3% | measured |
| APEX-Agents | 41.4% | estimated ± 4.9 pp, low confidence |
| APEX-Agents-AA | 47.1% | measured |
| BioMysteryBench (human-difficult) | 45.7% | estimated ± 6.8 pp, low confidence |
| DeepSearchQA | 88.1% | estimated ± 4.5 pp, high confidence |
| SWE-bench Verified | 76.8% | estimated ± 3.6 pp, high confidence |
| BFCL v4 | 76.0% | estimated ± 2.7 pp, medium confidence |
| CyberGym | 66.5% | estimated ± 5.3 pp, medium confidence |
| JobBench | 40.1% | estimated ± 7.3 pp, low confidence |
| SkillsBench | 59.9% | estimated ± 9.1 pp, low confidence |
| SWE-bench Pro | 55.1% | measured |
| WideResearch | 81.4% | estimated ± 2.1 pp, medium confidence |
| AA Agentic Index | 27.3% | measured |
| MMLU-Pro | 85.2% | estimated ± 4.4 pp, high confidence |
| AA-IFBench | 76.3% | measured |
| FailureSensorIQ | 68.2% | estimated ± 1.7 pp, high confidence |
| GPQA | 92.2% | measured |
| MMLU | 90.5% | estimated ± 0.9 pp, medium confidence |
| Vals GPQA Diamond | 92.7% | measured |
| Vals LiveCodeBench | 87.6% | measured |
| Vals MMLU-Pro | 89.5% | measured |
| MMMLU | 90.0% | estimated ± 0.7 pp, medium confidence |
| MMMU | 88.6% | estimated ± 0.8 pp, medium confidence |
| Chess Puzzles | 45.0% | measured |
| AA MMLU-Pro | 89.1% | estimated ± 0.7 pp, medium confidence |
| Gert Labs | 48.6% | measured |
| AA-AnalystAgent | 45.0% | measured |
| AA EnterpriseOps-Gym | 50.1% | measured |
| AA Harvey LAB v1.0 | 85.6% | estimated ± 1.2 pp, medium confidence |
| AA Tau3 Banking | 39.8% | estimated ± 2.0 pp, medium confidence |
| AA Terminal-Bench 2.1 | 84.0% | estimated ± 2.5 pp, high confidence |
| ApprenticeBench | 25.9% | estimated ± 5.0 pp, low confidence |
| ARC-AGI-1 | 92.5% | measured |
| ARC-AGI-2 | 72.1% | measured |
| BabaIsBench | 88.0% | measured |
| CursorBench 3.1 | 49.8% | measured |
| CursorBench 3.2 | 48.8% | measured |
| EBR-bench | 4.8% | measured |
| Terminal-Bench 3.0 | 18.2% | estimated ± 6.9 pp, medium confidence |
| FrontierCode 1.1 Main | 32.2% | estimated ± 1.6 pp, low confidence |
| FrontierMath v2 (Tier 4) | 26.8% | measured |
| FrontierMath v2 (Tiers 1-3) | 62.8% | measured |
| FrontierSWE v2 | 7.1% | estimated ± 8.3 pp, medium confidence |
| Furniture Assembly | 18.6% | estimated ± 7.4 pp, medium confidence |
| MaxxBench | 20.4% | estimated ± 4.3 pp, low confidence |
| MazeBench | 0.0% | measured |
| Mystery Game Puzzles | 32.0% | measured |
| NYT Connections | 92.6% | measured |
| OfficeQA Pro | 67.6% | estimated ± 6.8 pp, medium confidence |
| OSWorld-Verified | 78.4% | measured |
| ParseBench-1 | 73.2% | measured |
| Terminal-Bench 4.1 | 13.4% | estimated ± 10.9 pp, low confidence |
| Terminal-Bench Hard | 56.7% | estimated ± 3.9 pp, high confidence |
| Vals SWE-bench | 78.8% | measured |
| Terminal-Bench 2.1 (Vals AI) | 74.2% | measured |
| VulcanBench v3 | 85.4% | estimated ± 5.6 pp, low confidence |
| CWE-Bench-AA | 51.3% | estimated ± 8.1 pp, medium confidence |
| AA Cyber Index | 17.7% | estimated ± 9.2 pp, low confidence |
| DeepsecBench-AA | 16.7% | estimated ± 8.2 pp, medium confidence |
| Harvey LAB-AA v1.1 | 7.4% | estimated ± 4.0 pp, low confidence |
| HealthBench Professional (length-adjusted, Artificial Analysis) | 61.1% | estimated ± 4.3 pp, medium confidence |
| HealthBench Professional (raw, Artificial Analysis) | 59.6% | estimated ± 0.8 pp, low confidence |
| AA ITBench | 40.3% | measured |
| Bug Hunt Bench | 19.4% | estimated ± 9.0 pp, low confidence |
| Chartography (tools) | 85.9% | estimated ± 4.1 pp, low confidence |
| CursorBench 4.0 | 33.3% | estimated ± 2.0 pp, medium confidence |
| CWE-bench v1 | 57.4% | estimated ± 5.1 pp, medium confidence |
| DeepSWE | 55.8% | estimated ± 6.4 pp, medium confidence |
| DRACO | 86.9% | estimated ± 0.7 pp, low confidence |
| FrontierCode 2.0 | 14.7% | estimated ± 5.4 pp, low confidence |
| FrontierPhysics | 2.8% | estimated ± 5.0 pp, medium confidence |
| HLE w/o tools | 45.8% | estimated ± 2.7 pp, medium confidence |
| OSWorld 2.0 | 12.9% | estimated ± 3.6 pp, high confidence |
| OSWorld 2.1 | 60.2% | estimated ± 3.6 pp, high confidence |
| OSWorld 2.1 Offline | 35.8% | estimated ± 1.9 pp, low confidence |
| PostTrainBench v1.1 | 23.1% | estimated ± 3.7 pp, high confidence |
| ProgramBench | 50.6% | estimated ± 6.0 pp, low confidence |
| SpireClimb | 5.8% | estimated ± 7.2 pp, medium confidence |
| StealthMark | 35.4% | estimated ± 8.9 pp, low confidence |
| SWE-bench Multimodal | 57.0% | estimated ± 5.6 pp, low confidence |
| Toolathlon-Verified | 57.0% | estimated ± 2.3 pp, medium confidence |
| VulcanBench Frontier v4 | 58.9% | estimated ± 2.3 pp, low confidence |
| AA AutomationBench | 53.4% | estimated ± 9.9 pp, low confidence |
| GDP.pdf | 19.5% | estimated ± 4.2 pp, high confidence |
| AA Terminal-Bench 4.0 | 4.3% | estimated ± 11.0 pp, low confidence |
| ARC-AGI-3 | 0.0% | estimated ± 8.3 pp, low confidence |
| ARC-AGI-3 Provider Adapter | 2.9% | estimated ± 7.3 pp, medium confidence |
| cursorBench41 | 44.8% | estimated ± 3.7 pp, medium confidence |
| CWE-bench | 55.8% | estimated ± 4.6 pp, high confidence |
| MKA | 35.3% | estimated ± 4.6 pp, high confidence |
| SkillsBench official leaderboard (BenchFlow) | 48.2% | measured |
| BinaryAudit | 32.4% | estimated ± 12.0 pp, low confidence |
| CompileBench | 100.0% | estimated ± 10.0 pp, low confidence |
| OTelBench | 22.5% | estimated ± 6.6 pp, medium confidence |
| Vibe Code Bench | 48.7% | measured |
| FACTS Multimodal | 51.7% | estimated ± 4.0 pp, high confidence |
| Hermes Bench | 67.0% | estimated ± 2.9 pp, high confidence |
| Hermes Index | 37.2% | estimated ± 3.1 pp, high confidence |
| ParseBench-2 | 75.8% | measured |
| CharXiv | 84.2% | measured |
| CharXiv w/o tools | 88.5% | estimated ± 1.2 pp, high confidence |
| ExploitBench | 52.9% | estimated ± 7.2 pp, medium confidence |
| ExploitGym | 10.9% | estimated ± 2.7 pp, high confidence |
| AI2D_TEST | 93.5% | estimated ± 0.9 pp, low confidence |
| C-Eval | 95.2% | estimated ± 0.9 pp, low confidence |
| Claw-Eval | 63.7% | estimated ± 3.2 pp, high confidence |
| CountBench | 100.0% | estimated ± 2.9 pp, low confidence |
| DeepPlanning | 51.7% | estimated ± 13.1 pp, low confidence |
| DynaMath | 86.9% | estimated ± 0.8 pp, low confidence |
| ERQA | 68.8% | estimated ± 3.1 pp, medium confidence |
| HMMT Feb 2026 | 93.6% | estimated ± 1.5 pp, medium confidence |
| HMMT Nov 2025 | 95.2% | estimated ± 1.5 pp, low confidence |
| MathVision | 93.5% | estimated ± 3.0 pp, medium confidence |
| MCP Atlas | 83.6% | measured |
| MCP-Tasks | 79.7% | estimated ± 7.0 pp, low confidence |
| MLVU (M-Avg) | 87.0% | estimated ± 1.5 pp, high confidence |
| MMAnswerBench | 89.8% | estimated ± 2.4 pp, low confidence |
| MMLU-Redux | 97.4% | estimated ± 0.8 pp, low confidence |
| MMMU-Pro | 83.6% | measured |
| NL2Repo | 44.7% | estimated ± 4.7 pp, high confidence |
| PostTrainBench v1 | 19.7% | estimated ± 3.1 pp, medium confidence |
| QwenClawBench | 68.9% | estimated ± 2.6 pp, medium confidence |
| RealWorldQA | 85.1% | estimated ± 1.5 pp, high confidence |
| ScreenSpot Pro | 78.0% | estimated ± 6.8 pp, medium confidence |
| SimpleVQA | 72.9% | estimated ± 7.5 pp, medium confidence |
| SuperGPQA | 80.0% | estimated ± 7.2 pp, medium confidence |
| Terminal-Bench 2.0 | 75.6% | estimated ± 5.8 pp, medium confidence |
| Toolathlon | 56.5% | measured |
| V* | 100.0% | estimated ± 7.5 pp, low confidence |
| Video-MME (w/o subtitle) | 87.3% | estimated ± 1.3 pp, medium confidence |
| Video-MME (with subtitle) | 88.4% | estimated ± 2.9 pp, high confidence |
| VideoMMMU | 86.0% | estimated ± 1.5 pp, medium confidence |
| We-Math | 84.7% | estimated ± 7.4 pp, low confidence |
| AA LiveCodeBench | 90.9% | estimated ± 2.6 pp, medium confidence |
| HealthBench Hard | 30.2% | estimated ± 5.1 pp, low confidence |
| LiveCodeBench Pro | 79.8% | estimated ± 6.0 pp, low confidence |
| MedXpertQA (MM) | 81.7% | estimated ± 2.4 pp, high confidence |
| MedXpertQA (Text) | 62.7% | estimated ± 4.5 pp, medium confidence |
| MMLU-Pro (Arcee) | 87.8% | estimated ± 3.4 pp, low confidence |
| React Native Evals | 81.2% | estimated ± 3.0 pp, medium confidence |
| ResearchClawBench | 18.0% | measured |
| SWE-Rebench | 62.0% | estimated ± 6.5 pp, low confidence |
| PostTrainBench | 20.6% | estimated ± 3.7 pp, high confidence |
| DAYJOB | 1.3% | estimated ± 3.2 pp, high confidence |
| DAYJOB Healthcare | 0.7% | estimated ± 4.2 pp, high confidence |
| Finance Agent v2 | 57.9% | measured |
| BioMysteryBench (human-solvable) | 86.1% | estimated ± 0.6 pp, low confidence |
| FrontierCode 1.1 Extended | 54.3% | estimated ± 3.0 pp, low confidence |
| HealthBench (raw) | 51.2% | estimated ± 6.0 pp, medium confidence |
| HealthBench (length-adjusted) | 56.1% | estimated ± 2.5 pp, low confidence |
| HealthBench Professional | 55.4% | estimated ± 4.2 pp, low confidence |
| HealthBench Professional (raw) | 61.6% | estimated ± 4.5 pp, high confidence |
| HLE-Verified | 49.0% | estimated ± 2.0 pp, medium confidence |
| LABBench2 | 76.3% | estimated ± 1.7 pp, low confidence |
| ScarfBench | 41.2% | estimated ± 9.1 pp, low confidence |
| BlenderWorld (text-only) | 71.3% | estimated ± 2.8 pp, medium confidence |
| Cookie-Bench | 80.0% | estimated ± 1.3 pp, medium confidence |
| STEMBench | 2.6% | estimated ± 0.2 pp, medium confidence |
| SimpleQA | 34.7% | estimated ± 7.8 pp, low confidence |
| Agents' Last Exam | 31.2% | estimated ± 9.7 pp, medium confidence |
| BabyVision w/ Python | 94.9% | estimated ± 3.8 pp, low confidence |
| OpenHarmony Bench v1.0 | 58.9% | estimated ± 2.4 pp, high confidence |
| SEC-Bench Pro | 11.5% | estimated ± 7.5 pp, low confidence |
| BabyVision | 73.0% | estimated ± 9.1 pp, low confidence |
| MMVU | 80.5% | estimated ± 2.7 pp, low confidence |
| PerceptionBench | 62.6% | estimated ± 4.9 pp, low confidence |
| MRCRv2 | 77.3% | measured |
| MMLongBench-Doc | 60.9% | estimated ± 2.6 pp, low confidence |
| OCRBench V2 | 72.5% | estimated ± 3.3 pp, medium confidence |
| BioBasicsBench | 29.3% | estimated ± 2.2 pp, low confidence |
| LifeVal | 77.0% | estimated ± 2.3 pp, low confidence |
| MMMU-Pro w/ Python | 86.6% | estimated ± 2.0 pp, medium confidence |
| ExploitBench Internal Port | 0.9% | estimated ± 1.4 pp, medium confidence |