Google · model
Gemini 3.5 Flash-Lite benchmark scores
As of 2026-10-07, Gemini 3.5 Flash-Lite (Google) has measured scores on 19 benchmarks and estimated scores on 87 more.
| Benchmark | Score | Source |
|---|---|---|
| AA-GPQA Diamond | 83.8% | measured |
| AA-HLE | 18.8% | measured |
| AA-SciCode | 41.3% | measured |
| Artificial Analysis Intelligence Index | 22.2% | measured |
| CritPt | 0.0% | measured |
| GDPval-AA | 24.3% | measured |
| GPQA Diamond | 87.2% | estimated ± 1.7 pp, high confidence |
| HLE | 19.0% | estimated ± 9.5 pp, medium confidence |
| AA-LCR | 76.0% | measured |
| LiveCodeBench v6 | 87.0% | estimated ± 2.1 pp, high confidence |
| AA-Omniscience Accuracy | 29.5% | measured |
| SciCode | 34.9% | estimated ± 2.9 pp, low confidence |
| Terminal-Bench 2.1 | 54.0% | measured |
| BrowseComp | 83.8% | estimated ± 3.1 pp, high confidence |
| HLE w/ tools | 44.1% | estimated ± 6.1 pp, low confidence |
| LongBench v2 | 61.5% | estimated ± 5.9 pp, medium confidence |
| τ²-bench results | 89.6% | estimated ± 13.3 pp, low confidence |
| AA Coding Index | 49.3% | measured |
| AA-MMMU-Pro | 79.0% | measured |
| APEX-Agents | 22.4% | estimated ± 10.0 pp, low confidence |
| APEX-Agents-AA | 20.4% | estimated ± 9.1 pp, medium confidence |
| BioMysteryBench (human-difficult) | 34.6% | estimated ± 5.9 pp, low confidence |
| DeepSearchQA | 80.9% | estimated ± 4.5 pp, high confidence |
| SWE-bench Verified | 77.8% | estimated ± 2.5 pp, high confidence |
| AutomationBench | 28.4% | estimated ± 9.7 pp, low confidence |
| BFCL v4 | 73.9% | estimated ± 2.6 pp, high confidence |
| CyberGym | 57.8% | estimated ± 5.3 pp, low confidence |
| JobBench | 20.1% | estimated ± 8.6 pp, medium confidence |
| SWE-bench Pro | 54.2% | measured |
| WideResearch | 74.8% | estimated ± 9.1 pp, low confidence |
| AA Agentic Index | 15.9% | measured |
| MMLU-Pro | 85.7% | estimated ± 3.5 pp, high confidence |
| GPQA | 89.6% | estimated ± 1.6 pp, high confidence |
| Gert Labs | 63.3% | estimated ± 7.2 pp, medium confidence |
| AA-AnalystAgent | 16.9% | estimated ± 12.0 pp, low confidence |
| AA EnterpriseOps-Gym | 39.6% | estimated ± 5.9 pp, low confidence |
| AA Harvey LAB | 79.8% | estimated ± 1.2 pp, low confidence |
| ApprenticeBench | 7.2% | estimated ± 5.2 pp, low confidence |
| ARC-AGI-1 | 62.2% | estimated ± 5.4 pp, medium confidence |
| ARC-AGI-2 | 34.1% | estimated ± 12.6 pp, low confidence |
| CursorBench 3.1 | 22.6% | estimated ± 3.4 pp, low confidence |
| CursorBench 3.2 | 48.9% | estimated ± 3.6 pp, medium confidence |
| Terminal-Bench 3.0 | 2.0% | estimated ± 6.9 pp, low confidence |
| FrontierCode 1.1 Main | 20.7% | estimated ± 3.8 pp, medium confidence |
| FrontierSWE v2 | 4.8% | estimated ± 9.0 pp, low confidence |
| OfficeQA Pro | 51.4% | estimated ± 7.2 pp, medium confidence |
| OSWorld-Verified | 74.0% | measured |
| Vals GPQA Diamond | 83.8% | measured |
| Vals LiveCodeBench | 79.0% | measured |
| Vals MMLU-Pro | 85.8% | measured |
| Vals SWE-bench | 75.0% | measured |
| Terminal-Bench 2.1 (Vals AI) | 50.2% | measured |
| VulcanBench v3 | 80.5% | estimated ± 5.6 pp, low confidence |
| AA AutomationBench | 4.1% | estimated ± 7.0 pp, medium confidence |
| GDP.pdf | 9.4% | estimated ± 5.4 pp, medium confidence |
| AA Tau3 Banking | 22.0% | estimated ± 5.8 pp, medium confidence |
| AA Terminal-Bench 2.1 | 52.3% | estimated ± 3.5 pp, high confidence |
| AA Terminal-Bench 4.0 | 0.0% | estimated ± 8.9 pp, medium confidence |
| CursorBench 4.0 | 15.6% | estimated ± 1.6 pp, medium confidence |
| CWE-bench v1 | 34.2% | estimated ± 6.9 pp, low confidence |
| DeepSWE | 13.2% | estimated ± 6.4 pp, low confidence |
| HLE w/o tools | 32.6% | estimated ± 5.1 pp, medium confidence |
| OSWorld 2.0 | 8.1% | estimated ± 3.6 pp, high confidence |
| PostTrainBench v1.1 | 23.9% | estimated ± 2.7 pp, high confidence |
| Toolathlon-Verified | 41.9% | estimated ± 2.3 pp, low confidence |
| Vibe Code Bench | 21.7% | estimated ± 10.3 pp, low confidence |
| CharXiv | 85.6% | estimated ± 7.5 pp, low confidence |
| CharXiv w/o tools | 82.1% | estimated ± 2.4 pp, high confidence |
| ExploitGym | 0.0% | estimated ± 12.2 pp, low confidence |
| C-Eval | 92.0% | estimated ± 1.0 pp, medium confidence |
| Claw-Eval | 56.8% | estimated ± 13.6 pp, low confidence |
| MathVision | 89.7% | estimated ± 4.9 pp, high confidence |
| MCP Atlas | 64.6% | estimated ± 7.8 pp, low confidence |
| MMLU-Redux | 95.4% | estimated ± 1.1 pp, medium confidence |
| MMMU-Pro | 78.8% | estimated ± 4.4 pp, medium confidence |
| NL2Repo | 43.0% | estimated ± 4.7 pp, high confidence |
| QwenClawBench | 55.9% | estimated ± 6.0 pp, low confidence |
| ScreenSpot Pro | 78.0% | estimated ± 10.7 pp, low confidence |
| SuperGPQA | 74.2% | estimated ± 7.5 pp, medium confidence |
| Terminal-Bench 2.0 | 58.9% | estimated ± 5.9 pp, medium confidence |
| Toolathlon | 40.0% | estimated ± 6.3 pp, medium confidence |
| ERQA | 68.1% | estimated ± 3.9 pp, high confidence |
| HealthBench Hard | 11.8% | estimated ± 7.7 pp, low confidence |
| LiveCodeBench Pro | 79.0% | estimated ± 6.0 pp, low confidence |
| MedXpertQA (MM) | 76.5% | estimated ± 6.5 pp, low confidence |
| MMLU-Pro (Arcee) | 77.1% | estimated ± 4.0 pp, low confidence |
| React Native Evals | 73.2% | estimated ± 2.4 pp, medium confidence |
| ResearchClawBench | 15.0% | estimated ± 2.0 pp, high confidence |
| SWE-Rebench | 58.1% | estimated ± 8.9 pp, low confidence |
| ARC-AGI-3 | 1.7% | estimated ± 3.4 pp, medium confidence |
| BioMysteryBench (human-solvable) | 81.7% | estimated ± 0.6 pp, low confidence |
| DRACO | 71.5% | estimated ± 3.0 pp, medium confidence |
| FrontierCode 1.1 Extended | 22.4% | estimated ± 2.4 pp, low confidence |
| HealthBench (raw) | 26.5% | estimated ± 6.4 pp, low confidence |
| HealthBench (length-adjusted) | 53.9% | estimated ± 3.7 pp, low confidence |
| HealthBench Professional | 55.0% | estimated ± 1.3 pp, low confidence |
| HealthBench Professional (raw) | 38.7% | estimated ± 5.0 pp, low confidence |
| HLE-Verified | 15.3% | estimated ± 2.0 pp, low confidence |
| LABBench2 | 80.8% | estimated ± 1.8 pp, low confidence |
| MMMU | 86.4% | estimated ± 0.9 pp, medium confidence |
| LiveCodeBench | 97.1% | estimated ± 8.9 pp, low confidence |
| OpenHarmony Bench | 51.0% | estimated ± 2.4 pp, medium confidence |
| MMLU | 90.7% | estimated ± 1.1 pp, medium confidence |
| MMMU-Pro w/ Python | 81.1% | estimated ± 1.6 pp, high confidence |
| RealWorldQA | 87.1% | estimated ± 1.2 pp, medium confidence |
| OmniDocBench 1.5 | 92.0% | estimated ± 6.4 pp, low confidence |