Google · model
Gemini 2.5 Flash benchmark scores
As of 2026-10-07, Gemini 2.5 Flash (Google) has measured scores on 11 benchmarks and estimated scores on 50 more.
| Benchmark | Score | Source |
|---|---|---|
| AA-GPQA Diamond | 68.3% | measured |
| AA-HLE | 4.7% | measured |
| AIME26 | 95.5% | estimated ± 0.6 pp, medium confidence |
| Artificial Analysis Intelligence Index | 9.9% | measured |
| CritPt | 1.4% | measured |
| GDPval-AA | 0.0% | estimated ± 12.5 pp, low confidence |
| GPQA Diamond | 69.3% | estimated ± 2.2 pp, medium confidence |
| HLE | 29.0% | estimated ± 9.5 pp, medium confidence |
| IFBench | 55.8% | estimated ± 6.9 pp, low confidence |
| AA-LCR | 49.9% | measured |
| AA-Omniscience Accuracy | 26.1% | measured |
| IFEval | 88.1% | estimated ± 1.3 pp, high confidence |
| LongBench v2 | 0.0% | estimated ± 5.9 pp, low confidence |
| τ²-bench results | 14.9% | measured |
| AA-MMMU-Pro | 65.5% | measured |
| APEX-Agents-AA | 1.6% | estimated ± 11.4 pp, low confidence |
| BioMysteryBench (human-difficult) | 32.2% | estimated ± 5.9 pp, low confidence |
| BFCL v4 | 44.9% | estimated ± 9.8 pp, low confidence |
| AA Agentic Index | 0.0% | estimated ± 12.3 pp, low confidence |
| MMLU-Pro | 85.6% | estimated ± 3.5 pp, high confidence |
| FrontierMath v2 (Tier 4) | 4.2% | measured |
| FrontierMath v2 (Tiers 1-3) | 4.8% | measured |
| GPQA | 62.0% | estimated ± 3.2 pp, high confidence |
| AA-IFBench | 39.0% | measured |
| AA EnterpriseOps-Gym | 1.2% | estimated ± 3.6 pp, low confidence |
| ARC-AGI-1 | 2.3% | estimated ± 5.4 pp, low confidence |
| ARC-AGI-2 | 0.9% | estimated ± 14.2 pp, low confidence |
| OfficeQA Pro | 50.3% | estimated ± 7.2 pp, low confidence |
| Vals GPQA Diamond | 56.2% | estimated ± 7.4 pp, medium confidence |
| Vals MMLU-Pro | 84.3% | estimated ± 2.2 pp, high confidence |
| AA ITBench | 100.0% | estimated ± 3.6 pp, low confidence |
| HLE w/o tools | 3.7% | estimated ± 5.7 pp, low confidence |
| CharXiv | 70.4% | estimated ± 7.5 pp, low confidence |
| CharXiv w/o tools | 73.2% | estimated ± 2.4 pp, medium confidence |
| C-Eval | 83.2% | estimated ± 1.0 pp, low confidence |
| HMMT Nov 2025 | 93.3% | estimated ± 2.6 pp, medium confidence |
| MathVision | 81.5% | estimated ± 4.9 pp, high confidence |
| MCP-Tasks | 100.0% | estimated ± 8.2 pp, low confidence |
| MMAnswerBench | 82.9% | estimated ± 1.3 pp, medium confidence |
| MMLU-Redux | 94.8% | estimated ± 1.1 pp, medium confidence |
| MMMU-Pro | 72.5% | estimated ± 4.4 pp, medium confidence |
| ScreenSpot Pro | 65.1% | estimated ± 10.7 pp, low confidence |
| SuperGPQA | 71.1% | estimated ± 7.5 pp, medium confidence |
| ERQA | 5.6% | estimated ± 3.9 pp, medium confidence |
| HealthBench Hard | 0.1% | estimated ± 7.7 pp, low confidence |
| MedXpertQA (MM) | 20.1% | estimated ± 6.5 pp, low confidence |
| MMLU-Pro (Arcee) | 82.4% | estimated ± 4.0 pp, medium confidence |
| ARC-AGI-3 | 1.7% | estimated ± 3.4 pp, medium confidence |
| FrontierMath (legacy) | 5.9% | estimated ± 0.7 pp, low confidence |
| BioMysteryBench (human-solvable) | 81.7% | estimated ± 0.6 pp, low confidence |
| HealthBench (raw) | 7.4% | estimated ± 6.4 pp, low confidence |
| HealthBench (length-adjusted) | 53.9% | estimated ± 3.7 pp, low confidence |
| HealthBench Professional | 57.7% | estimated ± 4.8 pp, low confidence |
| HealthBench Professional (raw) | 21.6% | estimated ± 5.0 pp, low confidence |
| HLE-Verified | 15.3% | estimated ± 2.0 pp, low confidence |
| LABBench2 | 48.8% | estimated ± 1.8 pp, low confidence |
| MMMU | 76.2% | estimated ± 0.9 pp, high confidence |
| MMLU | 89.4% | estimated ± 1.1 pp, medium confidence |
| MMMU-Pro w/ Python | 72.3% | estimated ± 1.6 pp, high confidence |
| RealWorldQA | 79.3% | estimated ± 1.2 pp, low confidence |
| OmniDocBench 1.5 | 0.0% | estimated ± 6.4 pp, low confidence |