Xiaomi · model
MiMo-V2.5-Pro benchmark scores
As of 2026-10-07, MiMo-V2.5-Pro (Xiaomi) has measured scores on 25 benchmarks and estimated scores on 77 more.
| Benchmark | Score | Source |
|---|---|---|
| AA-GPQA Diamond | 86.6% | measured |
| AA-HLE | 35.7% | measured |
| AA-SciCode | 50.6% | measured |
| Artificial Analysis Intelligence Index | 26.0% | measured |
| CritPt | 4.0% | measured |
| GDPval-AA | 31.2% | measured |
| GPQA Diamond | 86.4% | estimated ± 1.7 pp, medium confidence |
| HLE | 48.0% | measured |
| IFBench | 78.9% | estimated ± 6.9 pp, medium confidence |
| AA-LCR | 79.7% | measured |
| LiveCodeBench v6 | 88.3% | estimated ± 2.1 pp, high confidence |
| AA-Omniscience Accuracy | 22.4% | measured |
| SciCode | 39.5% | estimated ± 2.9 pp, low confidence |
| Terminal-Bench 2.1 | 74.6% | estimated ± 6.9 pp, medium confidence |
| BrowseComp | 0.6% | estimated ± 3.4 pp, medium confidence |
| HLE w/ tools | 47.6% | estimated ± 6.1 pp, medium confidence |
| IFEval | 93.8% | estimated ± 1.3 pp, high confidence |
| LongBench v2 | 61.5% | estimated ± 5.9 pp, low confidence |
| VITA-Bench | 41.0% | estimated ± 10.7 pp, low confidence |
| τ²-bench results | 94.2% | measured |
| AA Coding Index | 60.2% | measured |
| APEX-Agents | 27.8% | estimated ± 10.0 pp, low confidence |
| APEX-Agents-AA | 2.4% | measured |
| BioMysteryBench (human-difficult) | 29.2% | estimated ± 5.9 pp, low confidence |
| DeepSearchQA | 85.9% | estimated ± 4.5 pp, high confidence |
| SWE-bench Verified | 80.3% | estimated ± 2.5 pp, high confidence |
| AutomationBench | 28.4% | estimated ± 9.7 pp, low confidence |
| BFCL v4 | 75.3% | estimated ± 2.6 pp, high confidence |
| CyberGym | 63.0% | estimated ± 5.3 pp, low confidence |
| JobBench | 0.0% | estimated ± 8.2 pp, low confidence |
| SWE-bench Pro | 57.2% | measured |
| τ³-bench results | 72.9% | measured |
| WideResearch | 78.9% | estimated ± 5.9 pp, low confidence |
| AA Agentic Index | 22.7% | measured |
| MMLU-Pro | 85.4% | estimated ± 2.5 pp, medium confidence |
| GPQA | 88.5% | estimated ± 1.6 pp, high confidence |
| AA-IFBench | 79.9% | measured |
| Gert Labs | 62.7% | measured |
| AA-AnalystAgent | 22.6% | estimated ± 12.0 pp, low confidence |
| AA EnterpriseOps-Gym | 45.1% | estimated ± 3.6 pp, medium confidence |
| AA Harvey LAB | 82.3% | estimated ± 1.2 pp, medium confidence |
| ApprenticeBench | 4.6% | estimated ± 3.1 pp, medium confidence |
| ARC-AGI-1 | 85.3% | estimated ± 3.7 pp, high confidence |
| ARC-AGI-2 | 28.9% | estimated ± 12.6 pp, low confidence |
| CursorBench 3.1 | 45.5% | estimated ± 3.4 pp, low confidence |
| CursorBench 3.2 | 54.1% | estimated ± 3.6 pp, high confidence |
| Terminal-Bench 3.0 | 4.4% | estimated ± 6.9 pp, low confidence |
| FrontierCode 1.1 Main | 29.4% | estimated ± 3.8 pp, high confidence |
| FrontierSWE v2 | 10.2% | estimated ± 9.0 pp, medium confidence |
| OSWorld-Verified | 72.5% | estimated ± 5.9 pp, medium confidence |
| Vals GPQA Diamond | 82.6% | measured |
| Vals LiveCodeBench | 81.4% | measured |
| Vals MMLU-Pro | 84.6% | measured |
| Vals SWE-bench | 74.0% | measured |
| Terminal-Bench 2.1 (Vals AI) | 57.3% | measured |
| VulcanBench v3 | 81.6% | estimated ± 5.6 pp, low confidence |
| AA AutomationBench | 6.5% | estimated ± 7.0 pp, medium confidence |
| GDP.pdf | 9.4% | estimated ± 5.4 pp, medium confidence |
| AA ITBench | 46.1% | estimated ± 3.6 pp, medium confidence |
| AA Tau3 Banking | 22.0% | estimated ± 5.8 pp, medium confidence |
| AA Terminal-Bench 2.1 | 73.0% | estimated ± 4.4 pp, high confidence |
| AA Terminal-Bench 4.0 | 10.6% | estimated ± 14.0 pp, low confidence |
| CursorBench 4.0 | 22.9% | estimated ± 1.6 pp, medium confidence |
| CWE-bench v1 | 41.0% | estimated ± 6.9 pp, medium confidence |
| DeepSWE | 33.4% | estimated ± 6.4 pp, low confidence |
| HLE w/o tools | 34.0% | measured |
| OSWorld 2.0 | 9.4% | estimated ± 4.5 pp, medium confidence |
| PostTrainBench v1.1 | 22.5% | estimated ± 2.7 pp, high confidence |
| Toolathlon-Verified | 51.7% | estimated ± 2.3 pp, low confidence |
| Vibe Code Bench | 33.9% | estimated ± 10.3 pp, low confidence |
| ExploitGym | 0.0% | estimated ± 12.2 pp, low confidence |
| C-Eval | 93.3% | estimated ± 0.8 pp, low confidence |
| Claw-Eval | 63.8% | measured |
| DeepPlanning | 35.4% | estimated ± 8.3 pp, low confidence |
| MCP Atlas | 68.5% | estimated ± 7.8 pp, low confidence |
| MCP-Tasks | 79.7% | estimated ± 3.6 pp, low confidence |
| MMLU-Redux | 93.9% | estimated ± 1.1 pp, medium confidence |
| NL2Repo | 42.5% | estimated ± 4.7 pp, high confidence |
| QwenClawBench | 60.7% | estimated ± 6.0 pp, low confidence |
| SuperGPQA | 67.2% | estimated ± 7.5 pp, medium confidence |
| Terminal-Bench 2.0 | 68.4% | measured |
| Toolathlon | 45.6% | estimated ± 6.0 pp, low confidence |
| HealthBench Hard | 20.0% | estimated ± 7.7 pp, low confidence |
| LiveCodeBench Pro | 81.5% | estimated ± 6.0 pp, low confidence |
| MedXpertQA (Text) | 51.0% | estimated ± 8.9 pp, low confidence |
| MMLU-Pro (Arcee) | 92.1% | estimated ± 4.0 pp, low confidence |
| React Native Evals | 71.6% | estimated ± 2.4 pp, medium confidence |
| ResearchClawBench | 17.4% | estimated ± 1.3 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 | 74.3% | estimated ± 3.0 pp, medium confidence |
| FrontierCode 1.1 Extended | 21.0% | estimated ± 2.4 pp, low confidence |
| HealthBench (raw) | 45.0% | estimated ± 6.4 pp, low confidence |
| HealthBench (length-adjusted) | 53.9% | estimated ± 3.7 pp, low confidence |
| HealthBench Professional | 51.7% | estimated ± 1.3 pp, low confidence |
| HealthBench Professional (raw) | 55.6% | estimated ± 5.0 pp, low confidence |
| HLE-Verified | 15.3% | estimated ± 2.0 pp, low confidence |
| LABBench2 | 80.1% | estimated ± 1.8 pp, low confidence |
| LiveCodeBench | 100.0% | estimated ± 8.9 pp, low confidence |
| OpenHarmony Bench | 56.4% | estimated ± 2.4 pp, high confidence |
| MMLU | 87.6% | estimated ± 1.1 pp, medium confidence |