Other · model
Hy4 flash benchmark scores
As of 2026-11-18, Hy4 flash (Other) has measured scores on 9 benchmarks and estimated scores on 74 more.
| Benchmark | Score | Source |
|---|---|---|
| AA-GPQA Diamond | 89.1% | estimated ± 7.1 pp, medium confidence |
| AA-HLE | 44.8% | estimated ± 6.5 pp, medium confidence |
| AA-SciCode | 56.4% | estimated ± 3.9 pp, medium confidence |
| Artificial Analysis Intelligence Index | 37.7% | estimated ± 10.7 pp, low confidence |
| CritPt | 18.4% | estimated ± 6.1 pp, medium confidence |
| GDPval-AA | 56.5% | estimated ± 4.2 pp, high confidence |
| GPQA Diamond | 92.6% | estimated ± 2.3 pp, high confidence |
| HLE | 55.2% | measured |
| AA-Omniscience Accuracy | 47.1% | estimated ± 9.0 pp, medium confidence |
| SciCode | 57.0% | estimated ± 7.6 pp, low confidence |
| BrowseComp | 89.9% | estimated ± 4.8 pp, medium confidence |
| HLE w/ tools | 55.2% | measured |
| τ²-bench results | 97.8% | estimated ± 3.1 pp, high confidence |
| AA Coding Index | 74.5% | estimated ± 2.5 pp, high confidence |
| AA-MMMU-Pro | 75.0% | estimated ± 1.5 pp, high confidence |
| APEX-Agents | 63.1% | measured |
| SWE-bench Verified | 85.6% | estimated ± 5.4 pp, medium confidence |
| AutomationBench | 52.2% | measured |
| CyberGym | 87.6% | estimated ± 4.6 pp, medium confidence |
| SWE-bench Pro | 65.5% | estimated ± 7.7 pp, medium confidence |
| WideResearch | 80.2% | estimated ± 4.8 pp, medium confidence |
| AA Agentic Index | 40.2% | estimated ± 9.2 pp, medium confidence |
| GPQA | 92.1% | estimated ± 1.9 pp, high confidence |
| Vals LiveCodeBench | 77.4% | estimated ± 6.1 pp, low confidence |
| Vals MMLU-Pro | 88.6% | estimated ± 2.4 pp, medium confidence |
| MMMU | 83.1% | estimated ± 1.7 pp, high confidence |
| AA Harvey LAB v1.0 | 89.5% | estimated ± 1.7 pp, medium confidence |
| AA Tau3 Banking | 50.2% | estimated ± 3.3 pp, low confidence |
| ARC-AGI-1 | 92.7% | estimated ± 11.6 pp, low confidence |
| CursorBench 3.2 | 17.6% | estimated ± 3.6 pp, low confidence |
| FrontierCode 1.1 Main | 46.9% | estimated ± 3.5 pp, low confidence |
| ParseBench-1 | 58.1% | estimated ± 8.4 pp, medium confidence |
| Terminal-Bench 4.1 | 16.0% | estimated ± 5.0 pp, medium confidence |
| Vals SWE-bench | 85.5% | estimated ± 10.6 pp, low confidence |
| CWE-Bench-AA | 56.2% | estimated ± 5.2 pp, medium confidence |
| AA Cyber Index | 46.3% | estimated ± 9.4 pp, medium confidence |
| Bug Hunt Bench | 26.4% | estimated ± 10.4 pp, low confidence |
| CursorBench 4.0 | 40.8% | estimated ± 4.7 pp, medium confidence |
| DeepSWE | 64.6% | measured |
| HLE w/o tools | 46.1% | estimated ± 6.0 pp, medium confidence |
| OSWorld 2.1 | 68.1% | measured |
| PostTrainBench v1.1 | 32.7% | estimated ± 4.5 pp, high confidence |
| ProgramBench | 16.0% | measured |
| Terminal-Bench 4.0 | 19.2% | measured |
| Terminal-Bench-Science 0.1 | 0.3% | estimated ± 14.9 pp, low confidence |
| Toolathlon-Verified | 70.3% | estimated ± 5.7 pp, medium confidence |
| VulcanBench Frontier v4 | 88.5% | estimated ± 1.2 pp, low confidence |
| AA AutomationBench | 41.9% | estimated ± 7.9 pp, low confidence |
| GDP.pdf | 14.7% | estimated ± 6.1 pp, low confidence |
| AA Terminal-Bench 4.0 | 18.0% | estimated ± 6.3 pp, low confidence |
| ARC-AGI-3 | 4.8% | estimated ± 8.3 pp, medium confidence |
| CWE-bench | 62.4% | estimated ± 11.0 pp, low confidence |
| Hermes Bench | 70.3% | estimated ± 1.8 pp, medium confidence |
| Hermes Index | 38.2% | estimated ± 9.3 pp, low confidence |
| ParseBench-2 | 65.9% | estimated ± 5.0 pp, medium confidence |
| CharXiv | 76.5% | estimated ± 5.1 pp, medium confidence |
| CharXiv w/o tools | 79.5% | estimated ± 2.1 pp, high confidence |
| AI2D_TEST | 91.2% | estimated ± 2.0 pp, medium confidence |
| CountBench | 94.9% | estimated ± 3.0 pp, high confidence |
| DynaMath | 84.5% | estimated ± 5.3 pp, low confidence |
| ERQA | 55.9% | estimated ± 6.4 pp, medium confidence |
| MathVision | 84.6% | estimated ± 2.9 pp, high confidence |
| MCP Atlas | 84.7% | estimated ± 6.6 pp, low confidence |
| MLVU (M-Avg) | 84.9% | estimated ± 2.2 pp, high confidence |
| MMLU-Redux | 96.9% | estimated ± 2.1 pp, medium confidence |
| MMMU-Pro | 75.8% | measured |
| NL2Repo | 58.5% | estimated ± 6.0 pp, medium confidence |
| RealWorldQA | 83.5% | estimated ± 2.9 pp, high confidence |
| ScreenSpot Pro | 74.3% | estimated ± 10.1 pp, low confidence |
| SimpleVQA | 63.7% | estimated ± 7.5 pp, medium confidence |
| SuperGPQA | 94.1% | estimated ± 7.2 pp, low confidence |
| Video-MME (w/o subtitle) | 83.3% | estimated ± 1.4 pp, medium confidence |
| Video-MME (with subtitle) | 85.8% | estimated ± 3.0 pp, high confidence |
| VideoMMMU | 81.6% | estimated ± 5.1 pp, medium confidence |
| We-Math | 81.3% | estimated ± 7.4 pp, low confidence |
| MedXpertQA (MM) | 65.1% | estimated ± 2.4 pp, high confidence |
| ScarfBench | 64.4% | estimated ± 3.8 pp, medium confidence |
| MMVU | 75.8% | estimated ± 2.7 pp, medium confidence |
| PerceptionBench | 43.9% | estimated ± 2.6 pp, low confidence |
| SWE-Marathon | 34.4% | estimated ± 12.9 pp, low confidence |
| MMMU-Pro w/ Python | 79.2% | estimated ± 2.6 pp, high confidence |
| OCRBench V2 | 71.9% | estimated ± 4.4 pp, medium confidence |
| SWE-Atlas Codebase QnA | 58.8% | estimated ± 4.8 pp, high confidence |