Meta · model
Muse Spark 1.3 benchmark scores
As of 2026-11-22, Muse Spark 1.3 (Meta) has measured scores on 36 benchmarks and estimated scores on 119 more.
| Benchmark | Score | Source |
|---|---|---|
| AA-GPQA Diamond | 93.5% | measured |
| AA-HLE | 48.7% | measured |
| AA-SciCode | 58.8% | measured |
| AIME26 | 99.2% | estimated ± 1.2 pp, low confidence |
| Artificial Analysis Intelligence Index | 48.1% | measured |
| CritPt | 24.9% | measured |
| GDPval-AA | 59.0% | measured |
| GPQA Diamond | 93.9% | estimated ± 2.4 pp, high confidence |
| HLE | 56.6% | estimated ± 8.5 pp, medium confidence |
| AA-LCR | 83.0% | measured |
| LiveCodeBench v6 | 88.8% | estimated ± 4.7 pp, medium confidence |
| AA-Omniscience Accuracy | 43.6% | measured |
| SciCode | 58.4% | estimated ± 3.1 pp, high confidence |
| Terminal-Bench 2.1 | 88.8% | measured |
| BrowseComp | 91.7% | estimated ± 1.9 pp, high confidence |
| HLE w/ tools | 62.3% | estimated ± 5.3 pp, medium confidence |
| LongBench v2 | 68.4% | estimated ± 5.5 pp, medium confidence |
| τ²-bench results | 92.8% | estimated ± 6.5 pp, low confidence |
| AA Coding Index | 75.8% | measured |
| APEX-Agents | 38.0% | estimated ± 7.4 pp, low confidence |
| APEX-Agents-AA | 38.8% | estimated ± 2.9 pp, low confidence |
| BioMysteryBench (human-difficult) | 49.0% | estimated ± 6.8 pp, low confidence |
| DeepSearchQA | 89.4% | measured |
| SWE-bench Verified | 90.7% | estimated ± 5.1 pp, low confidence |
| AutomationBench | 49.4% | measured |
| BFCL v4 | 77.2% | estimated ± 2.7 pp, medium confidence |
| CyberGym | 85.4% | estimated ± 3.1 pp, low confidence |
| JobBench | 64.9% | measured |
| SWE-bench Pro | 67.2% | estimated ± 4.2 pp, high confidence |
| WideResearch | 84.5% | estimated ± 1.2 pp, low confidence |
| AA Agentic Index | 55.7% | measured |
| MMLU-Pro | 85.5% | estimated ± 5.0 pp, medium confidence |
| FailureSensorIQ | 67.7% | estimated ± 1.7 pp, high confidence |
| GPQA | 93.6% | estimated ± 6.1 pp, medium confidence |
| MMLU | 90.5% | estimated ± 0.9 pp, medium confidence |
| Vals GPQA Diamond | 92.6% | estimated ± 3.2 pp, high confidence |
| Vals LiveCodeBench | 84.7% | estimated ± 3.0 pp, high confidence |
| Vals MMLU-Pro | 87.8% | estimated ± 2.9 pp, high confidence |
| MMMLU | 91.3% | estimated ± 0.7 pp, low confidence |
| Chess Puzzles | 37.0% | measured |
| AA MMLU-Pro | 87.7% | estimated ± 0.8 pp, high confidence |
| Gert Labs | 57.0% | measured |
| AA-AnalystAgent | 44.8% | estimated ± 5.9 pp, low confidence |
| AA EnterpriseOps-Gym | 46.8% | estimated ± 1.2 pp, medium confidence |
| AA Harvey LAB v1.0 | 93.7% | estimated ± 1.2 pp, medium confidence |
| AA Tau3 Banking | 50.5% | measured |
| AA Terminal-Bench 2.1 | 83.4% | estimated ± 2.5 pp, high confidence |
| ApprenticeBench | 19.0% | measured |
| ARC-AGI-1 | 93.4% | estimated ± 7.8 pp, low confidence |
| ARC-AGI-2 | 78.4% | estimated ± 14.4 pp, low confidence |
| BabaIsBench | 46.0% | measured |
| CursorBench 3.1 | 65.7% | estimated ± 3.4 pp, medium confidence |
| CursorBench 3.2 | 67.2% | estimated ± 2.0 pp, high confidence |
| EBR-bench | 14.5% | estimated ± 10.4 pp, low confidence |
| Terminal-Bench 3.0 | 16.9% | estimated ± 3.2 pp, high confidence |
| FrontierCode 1.1 Main | 52.6% | estimated ± 3.5 pp, low confidence |
| FrontierMath v2 (Tier 4) | 46.3% | measured |
| FrontierMath v2 (Tiers 1-3) | 74.0% | measured |
| FrontierSWE v2 | 25.8% | measured |
| Furniture Assembly | 33.2% | estimated ± 9.3 pp, medium confidence |
| MaxxBench | 49.7% | estimated ± 5.7 pp, medium confidence |
| MazeBench | 0.0% | measured |
| MirrorCode | 54.4% | measured |
| Mystery Game Puzzles | 25.0% | measured |
| NYT Connections | 85.1% | measured |
| OSWorld-Verified | 79.5% | estimated ± 2.3 pp, medium confidence |
| Terminal-Bench 4.1 | 14.9% | measured |
| Terminal-Bench Hard | 56.6% | estimated ± 2.8 pp, medium confidence |
| Vals SWE-bench | 94.7% | estimated ± 3.7 pp, high confidence |
| Terminal-Bench 2.1 (Vals AI) | 72.3% | measured |
| VulcanBench v3 | 87.6% | estimated ± 3.9 pp, medium confidence |
| CWE-Bench-AA | 60.1% | estimated ± 4.5 pp, medium confidence |
| AA Cyber Index | 31.7% | estimated ± 9.2 pp, low confidence |
| DeepsecBench-AA | 20.8% | estimated ± 8.2 pp, medium confidence |
| HealthBench Professional (length-adjusted, Artificial Analysis) | 60.5% | estimated ± 4.3 pp, medium confidence |
| HealthBench Professional (raw, Artificial Analysis) | 68.0% | estimated ± 0.8 pp, medium confidence |
| AA ITBench | 45.3% | estimated ± 3.2 pp, high confidence |
| Bug Hunt Bench | 32.2% | measured |
| CursorBench 4.0 | 41.6% | measured |
| CWE-bench v1 | 55.0% | measured |
| DeepSWE | 75.4% | measured |
| DRACO | 86.9% | estimated ± 0.7 pp, low confidence |
| FrontierCode 2.0 | 27.1% | estimated ± 4.7 pp, medium confidence |
| FrontierPhysics | 1.8% | estimated ± 4.7 pp, high confidence |
| HLE w/o tools | 54.4% | estimated ± 4.9 pp, high confidence |
| OSWorld 2.0 | 66.9% | measured |
| OSWorld 2.1 | 71.0% | estimated ± 3.6 pp, high confidence |
| OSWorld 2.1 Offline | 77.7% | estimated ± 1.9 pp, medium confidence |
| PostTrainBench v1.1 | 35.6% | estimated ± 1.7 pp, low confidence |
| SpireClimb | 5.8% | estimated ± 7.2 pp, medium confidence |
| StealthMark | 39.5% | measured |
| Terminal-Bench 4.0 | 22.1% | estimated ± 4.5 pp, high confidence |
| Terminal-Bench-Science 0.1 | 0.0% | estimated ± 10.7 pp, low confidence |
| Toolathlon-Verified | 79.7% | estimated ± 1.0 pp, medium confidence |
| VulcanBench Frontier v4 | 88.6% | estimated ± 1.8 pp, high confidence |
| AA AutomationBench | 58.0% | estimated ± 9.9 pp, low confidence |
| GDP.pdf | 21.4% | estimated ± 4.1 pp, medium confidence |
| AA Terminal-Bench 4.0 | 14.5% | estimated ± 6.2 pp, medium confidence |
| ARC-AGI-3 | 3.4% | estimated ± 9.8 pp, medium confidence |
| ARC-AGI-3 Provider Adapter | 46.6% | estimated ± 7.3 pp, medium confidence |
| cursorBench41 | 41.9% | measured |
| CWE-bench | 54.3% | estimated ± 5.8 pp, low confidence |
| MKA | 38.5% | estimated ± 4.6 pp, high confidence |
| SkillsBench official leaderboard (BenchFlow) | 55.1% | estimated ± 6.1 pp, medium confidence |
| BinaryAudit | 35.8% | estimated ± 12.1 pp, low confidence |
| CompileBench | 95.5% | estimated ± 10.0 pp, low confidence |
| OTelBench | 27.5% | estimated ± 4.1 pp, low confidence |
| Vibe Code Bench | 73.9% | estimated ± 10.0 pp, low confidence |
| FACTS Multimodal | 47.8% | estimated ± 4.0 pp, high confidence |
| Hermes Bench | 72.7% | estimated ± 2.4 pp, low confidence |
| Hermes Index | 37.2% | estimated ± 3.1 pp, high confidence |
| ExploitBench | 43.9% | estimated ± 7.2 pp, medium confidence |
| ExploitGym | 14.5% | estimated ± 2.7 pp, high confidence |
| C-Eval | 94.7% | estimated ± 1.2 pp, low confidence |
| Claw-Eval | 78.6% | estimated ± 4.5 pp, medium confidence |
| HMMT Feb 2026 | 100.0% | estimated ± 1.5 pp, low confidence |
| HMMT Nov 2025 | 95.6% | estimated ± 1.5 pp, low confidence |
| MCP Atlas | 78.9% | estimated ± 5.8 pp, low confidence |
| MMAnswerBench | 92.7% | estimated ± 2.4 pp, low confidence |
| MMLU-Redux | 96.4% | estimated ± 0.8 pp, low confidence |
| NL2Repo | 56.3% | estimated ± 4.7 pp, medium confidence |
| SuperGPQA | 87.8% | estimated ± 9.3 pp, medium confidence |
| Terminal-Bench 2.0 | 74.5% | estimated ± 5.8 pp, medium confidence |
| Toolathlon | 55.6% | estimated ± 1.7 pp, medium confidence |
| HealthBench Hard | 33.4% | estimated ± 8.6 pp, low confidence |
| React Native Evals | 83.2% | estimated ± 3.2 pp, medium confidence |
| ResearchClawBench | 21.6% | estimated ± 1.9 pp, medium confidence |
| PostTrainBench | 33.1% | estimated ± 2.4 pp, high confidence |
| DAYJOB | 0.4% | estimated ± 3.2 pp, high confidence |
| DAYJOB Healthcare | 0.7% | estimated ± 4.2 pp, high confidence |
| Finance Agent v2 | 56.6% | estimated ± 2.4 pp, medium confidence |
| BioMysteryBench (human-solvable) | 89.4% | estimated ± 0.6 pp, medium confidence |
| FrontierCode 1.1 Extended | 59.5% | estimated ± 3.0 pp, medium confidence |
| HealthBench (raw) | 56.3% | estimated ± 6.0 pp, medium confidence |
| HealthBench (length-adjusted) | 56.9% | estimated ± 3.4 pp, medium confidence |
| HealthBench Professional | 60.0% | estimated ± 4.2 pp, medium confidence |
| HealthBench Professional (raw) | 66.5% | estimated ± 4.5 pp, high confidence |
| HLE-Verified | 54.1% | estimated ± 2.0 pp, medium confidence |
| LABBench2 | 82.7% | estimated ± 1.8 pp, low confidence |
| ScarfBench | 69.5% | estimated ± 3.8 pp, medium confidence |
| BlenderWorld (text-only) | 70.3% | estimated ± 2.8 pp, medium confidence |
| Cookie-Bench | 77.7% | estimated ± 1.3 pp, medium confidence |
| STEMBench | 2.5% | estimated ± 0.2 pp, medium confidence |
| SimpleQA | 50.5% | estimated ± 4.3 pp, low confidence |
| Agents' Last Exam | 33.6% | estimated ± 9.7 pp, medium confidence |
| OpenHarmony Bench v1.0 | 58.9% | estimated ± 2.4 pp, high confidence |
| OpenHarmony Bench v1.2 | 63.8% | estimated ± 11.0 pp, low confidence |
| SEC-Bench Pro | 46.0% | estimated ± 9.3 pp, medium confidence |
| MRCRv2 | 85.9% | estimated ± 11.3 pp, low confidence |
| MMLongBench-Doc | 61.5% | estimated ± 2.6 pp, low confidence |
| SWE-Marathon | 50.9% | estimated ± 2.1 pp, medium confidence |
| BioBasicsBench | 41.8% | estimated ± 2.8 pp, medium confidence |
| LifeVal | 75.2% | estimated ± 3.8 pp, medium confidence |
| ExploitBench Internal Port | 0.3% | estimated ± 1.4 pp, medium confidence |
| SWE-Atlas Codebase QnA | 59.4% | measured |