Meta · model
Muse Spark 1.1 benchmark scores
As of 2026-11-22, Muse Spark 1.1 (Meta) has measured scores on 35 benchmarks and estimated scores on 135 more.
| Benchmark | Score | Source |
|---|---|---|
| AA-GPQA Diamond | 89.8% | measured |
| AA-HLE | 46.2% | measured |
| AA-SciCode | 58.8% | measured |
| Artificial Analysis Intelligence Index | 33.7% | measured |
| CritPt | 15.1% | measured |
| GDPval-AA | 35.6% | measured |
| GPQA Diamond | 93.6% | estimated ± 1.2 pp, medium confidence |
| HLE | 62.1% | measured |
| AA-LCR | 77.7% | measured |
| LiveCodeBench v6 | 89.1% | estimated ± 2.2 pp, medium confidence |
| AA-Omniscience Accuracy | 52.1% | measured |
| SciCode | 48.5% | estimated ± 2.9 pp, medium confidence |
| Terminal-Bench 2.1 | 80.0% | measured |
| BrowseComp | 82.9% | estimated ± 1.9 pp, high confidence |
| HLE w/ tools | 50.0% | estimated ± 4.9 pp, medium confidence |
| LongBench v2 | 63.8% | estimated ± 5.5 pp, medium confidence |
| VITA-Bench | 51.3% | estimated ± 9.9 pp, low confidence |
| AA Coding Index | 71.3% | measured |
| AA-MMMU-Pro | 80.0% | estimated ± 4.9 pp, medium confidence |
| APEX-Agents | 27.6% | estimated ± 9.4 pp, low confidence |
| APEX-Agents-AA | 35.2% | estimated ± 2.9 pp, medium confidence |
| BioMysteryBench (human-difficult) | 46.6% | estimated ± 6.8 pp, low confidence |
| DeepSearchQA | 84.9% | measured |
| SWE-bench Verified | 82.0% | estimated ± 3.6 pp, high confidence |
| AutomationBench | 11.0% | estimated ± 10.4 pp, low confidence |
| BFCL v4 | 75.2% | estimated ± 2.7 pp, medium confidence |
| CyberGym | 59.0% | measured |
| JobBench | 54.7% | measured |
| SkillsBench | 62.1% | estimated ± 9.1 pp, low confidence |
| SWE-bench Pro | 61.5% | measured |
| WideResearch | 82.3% | estimated ± 1.2 pp, medium confidence |
| AA Agentic Index | 27.5% | measured |
| MMLU-Pro | 85.9% | estimated ± 2.5 pp, low confidence |
| FailureSensorIQ | 67.7% | estimated ± 1.7 pp, high confidence |
| GPQA | 93.6% | estimated ± 1.2 pp, medium confidence |
| MMLU | 90.5% | estimated ± 0.9 pp, medium confidence |
| Vals GPQA Diamond | 91.2% | measured |
| Vals LiveCodeBench | 85.9% | measured |
| Vals MMLU-Pro | 88.7% | measured |
| MMMLU | 90.3% | estimated ± 0.7 pp, low confidence |
| MMMU | 89.4% | estimated ± 3.7 pp, low confidence |
| Chess Puzzles | 34.0% | measured |
| AA MMLU-Pro | 89.1% | estimated ± 0.7 pp, high confidence |
| Gert Labs | 56.1% | measured |
| AA-AnalystAgent | 55.7% | estimated ± 3.7 pp, low confidence |
| AA EnterpriseOps-Gym | 46.4% | estimated ± 1.2 pp, medium confidence |
| AA Harvey LAB v1.0 | 85.8% | estimated ± 1.2 pp, medium confidence |
| AA Tau3 Banking | 31.8% | estimated ± 3.7 pp, high confidence |
| AA Terminal-Bench 2.1 | 82.5% | estimated ± 2.5 pp, high confidence |
| ApprenticeBench | 4.1% | estimated ± 4.0 pp, medium confidence |
| ARC-AGI-1 | 96.3% | estimated ± 3.9 pp, high confidence |
| ARC-AGI-2 | 82.0% | estimated ± 13.0 pp, low confidence |
| CursorBench 3.1 | 65.7% | estimated ± 3.4 pp, medium confidence |
| CursorBench 3.2 | 60.0% | estimated ± 3.6 pp, high confidence |
| EBR-bench | 19.5% | estimated ± 12.7 pp, low confidence |
| Terminal-Bench 3.0 | 12.7% | estimated ± 6.9 pp, medium confidence |
| FrontierCode 1.1 Main | 40.3% | estimated ± 3.5 pp, low confidence |
| FrontierSWE v2 | 18.2% | estimated ± 8.3 pp, medium confidence |
| Furniture Assembly | 32.3% | estimated ± 14.1 pp, low confidence |
| MaxxBench | 49.1% | estimated ± 10.3 pp, low confidence |
| MazeBench | 1.4% | estimated ± 2.7 pp, high confidence |
| Mystery Game Puzzles | 25.0% | estimated ± 10.3 pp, low confidence |
| NYT Connections | 84.9% | measured |
| OSWorld-Verified | 80.8% | measured |
| ParseBench-1 | 66.4% | estimated ± 8.6 pp, low confidence |
| Terminal-Bench 4.1 | 17.1% | estimated ± 10.9 pp, low confidence |
| Terminal-Bench Hard | 53.0% | estimated ± 3.9 pp, high confidence |
| Vals SWE-bench | 82.0% | measured |
| Terminal-Bench 2.1 (Vals AI) | 69.3% | measured |
| VulcanBench v3 | 84.2% | estimated ± 5.6 pp, low confidence |
| CWE-Bench-AA | 45.0% | estimated ± 4.5 pp, medium confidence |
| AA Cyber Index | 17.9% | estimated ± 9.2 pp, low confidence |
| DeepsecBench-AA | 20.3% | estimated ± 8.2 pp, medium confidence |
| HealthBench Professional (length-adjusted, Artificial Analysis) | 56.1% | estimated ± 4.3 pp, low confidence |
| HealthBench Professional (raw, Artificial Analysis) | 61.7% | estimated ± 0.8 pp, low confidence |
| AA ITBench | 41.7% | estimated ± 4.0 pp, high confidence |
| Bug Hunt Bench | 21.1% | estimated ± 9.0 pp, low confidence |
| CursorBench 4.0 | 37.1% | estimated ± 2.0 pp, medium confidence |
| CWE-bench v1 | 46.2% | estimated ± 4.5 pp, low confidence |
| DeepSWE | 53.3% | measured |
| DRACO | 79.0% | estimated ± 2.0 pp, medium confidence |
| FrontierCode 2.0 | 21.7% | estimated ± 6.7 pp, medium confidence |
| FrontierPhysics | 0.3% | estimated ± 6.1 pp, medium confidence |
| HLE w/o tools | 52.2% | measured |
| OSWorld 2.0 | 14.2% | measured |
| OSWorld 2.1 | 70.3% | estimated ± 3.6 pp, high confidence |
| OSWorld 2.1 Offline | 17.0% | estimated ± 1.9 pp, low confidence |
| PostTrainBench v1.1 | 25.8% | estimated ± 3.7 pp, high confidence |
| SpireClimb | 5.8% | estimated ± 7.2 pp, medium confidence |
| StealthMark | 44.9% | estimated ± 9.7 pp, low confidence |
| Toolathlon-Verified | 69.8% | estimated ± 1.0 pp, low confidence |
| VulcanBench Frontier v4 | 83.3% | estimated ± 4.0 pp, low confidence |
| AA AutomationBench | 8.9% | estimated ± 4.8 pp, low confidence |
| GDP.pdf | 16.9% | estimated ± 4.1 pp, medium confidence |
| AA Terminal-Bench 4.0 | 4.3% | estimated ± 11.0 pp, low confidence |
| ARC-AGI-3 | 14.3% | estimated ± 7.2 pp, medium confidence |
| ARC-AGI-3 Provider Adapter | 9.5% | estimated ± 7.3 pp, medium confidence |
| cursorBench41 | 45.1% | estimated ± 3.7 pp, high confidence |
| CWE-bench | 50.7% | estimated ± 6.2 pp, medium confidence |
| MKA | 4.9% | estimated ± 3.1 pp, medium confidence |
| SkillsBench official leaderboard (BenchFlow) | 54.7% | estimated ± 6.1 pp, medium confidence |
| BinaryAudit | 35.8% | estimated ± 12.1 pp, low confidence |
| OTelBench | 22.5% | estimated ± 6.6 pp, medium confidence |
| Vibe Code Bench | 57.6% | estimated ± 10.0 pp, low confidence |
| FACTS Multimodal | 47.7% | estimated ± 4.0 pp, high confidence |
| Hermes Bench | 33.8% | estimated ± 1.5 pp, low confidence |
| Hermes Index | 3.7% | estimated ± 5.5 pp, low confidence |
| CharXiv | 88.4% | measured |
| CharXiv w/o tools | 86.8% | estimated ± 2.0 pp, medium confidence |
| ExploitBench | 37.7% | estimated ± 13.4 pp, low confidence |
| ExploitGym | 0.8% | measured |
| AI2D_TEST | 93.5% | estimated ± 0.9 pp, low confidence |
| C-Eval | 95.7% | estimated ± 1.0 pp, low confidence |
| Claw-Eval | 73.5% | estimated ± 4.5 pp, medium confidence |
| CountBench | 98.1% | estimated ± 3.7 pp, low confidence |
| DeepPlanning | 53.8% | estimated ± 13.1 pp, low confidence |
| DynaMath | 87.7% | estimated ± 0.8 pp, low confidence |
| ERQA | 70.3% | estimated ± 5.6 pp, medium confidence |
| MathVision | 93.3% | estimated ± 0.9 pp, high confidence |
| MCP Atlas | 88.1% | measured |
| MLVU (M-Avg) | 88.4% | estimated ± 1.5 pp, high confidence |
| MMLU-Redux | 97.5% | estimated ± 0.8 pp, low confidence |
| MMMU-Pro | 81.9% | estimated ± 2.5 pp, medium confidence |
| NL2Repo | 46.9% | estimated ± 4.7 pp, high confidence |
| PostTrainBench v1 | 26.8% | estimated ± 3.6 pp, high confidence |
| QwenClawBench | 73.0% | estimated ± 2.6 pp, medium confidence |
| RealWorldQA | 86.7% | estimated ± 1.5 pp, high confidence |
| ScreenSpot Pro | 83.4% | estimated ± 6.8 pp, medium confidence |
| SuperGPQA | 100.0% | estimated ± 7.2 pp, low confidence |
| Terminal-Bench 2.0 | 72.6% | estimated ± 5.8 pp, medium confidence |
| Toolathlon | 75.6% | measured |
| Video-MME (with subtitle) | 90.1% | estimated ± 2.9 pp, high confidence |
| VideoMMMU | 86.9% | estimated ± 1.5 pp, medium confidence |
| We-Math | 84.6% | estimated ± 7.9 pp, low confidence |
| AA LiveCodeBench | 92.8% | estimated ± 2.6 pp, medium confidence |
| HealthBench Hard | 31.3% | estimated ± 8.6 pp, low confidence |
| LiveCodeBench Pro | 84.9% | estimated ± 6.0 pp, low confidence |
| MedXpertQA (MM) | 78.4% | estimated ± 6.4 pp, low confidence |
| MedXpertQA (Text) | 68.9% | estimated ± 8.9 pp, low confidence |
| React Native Evals | 82.1% | estimated ± 3.2 pp, high confidence |
| ResearchClawBench | 20.1% | estimated ± 1.8 pp, medium confidence |
| PostTrainBench | 23.5% | estimated ± 3.7 pp, high confidence |
| DAYJOB | 0.8% | estimated ± 6.3 pp, medium confidence |
| DAYJOB Healthcare | 0.4% | estimated ± 6.1 pp, low confidence |
| Finance Agent v2 | 57.2% | measured |
| BioMysteryBench (human-solvable) | 87.6% | estimated ± 0.6 pp, medium confidence |
| FrontierCode 1.1 Extended | 56.5% | estimated ± 3.0 pp, low confidence |
| HealthBench (raw) | 54.2% | estimated ± 6.0 pp, medium confidence |
| HealthBench (length-adjusted) | 52.4% | estimated ± 2.2 pp, medium confidence |
| HealthBench Professional | 59.3% | measured |
| HealthBench Professional (raw) | 64.5% | estimated ± 4.5 pp, high confidence |
| HLE-Verified | 17.2% | estimated ± 2.0 pp, low confidence |
| LABBench2 | 82.4% | estimated ± 1.8 pp, low confidence |
| ScarfBench | 61.8% | estimated ± 3.8 pp, low confidence |
| BlenderWorld (text-only) | 70.6% | estimated ± 4.0 pp, medium confidence |
| Cookie-Bench | 81.2% | estimated ± 2.5 pp, medium confidence |
| STEMBench | 2.6% | estimated ± 0.2 pp, medium confidence |
| SimpleQA | 33.5% | estimated ± 7.8 pp, low confidence |
| Agents' Last Exam | 20.6% | estimated ± 3.3 pp, low confidence |
| OpenHarmony Bench v1.0 | 58.9% | estimated ± 2.4 pp, high confidence |
| SEC-Bench Pro | 46.8% | estimated ± 12.3 pp, low confidence |
| BabyVision | 76.3% | measured |
| MMVU | 79.9% | estimated ± 3.6 pp, medium confidence |
| MRCRv2 | 85.4% | estimated ± 11.3 pp, low confidence |
| MMLongBench-Doc | 61.3% | estimated ± 2.6 pp, low confidence |
| BioBasicsBench | 35.7% | estimated ± 2.8 pp, medium confidence |
| LifeVal | 67.4% | estimated ± 5.0 pp, medium confidence |
| ExploitBench Internal Port | 0.2% | estimated ± 1.4 pp, low confidence |
| MathVision w/ Python | 94.9% | estimated ± 1.0 pp, low confidence |
| SWE-Atlas Codebase QnA | 50.2% | estimated ± 4.8 pp, high confidence |