Meta · model
Muse Glimmer 30B benchmark scores
As of 2026-10-07, Muse Glimmer 30B (Meta) has measured scores on 31 benchmarks and estimated scores on 95 more.
| Benchmark | Score | Source |
|---|---|---|
| AA-GPQA Diamond | 83.5% | measured |
| AA-HLE | 22.0% | measured |
| AA-SciCode | 44.9% | measured |
| AIME26 | 94.7% | measured |
| Artificial Analysis Intelligence Index | 17.5% | measured |
| CritPt | 2.6% | measured |
| GDPval-AA | 14.5% | measured |
| GPQA Diamond | 82.5% | estimated ± 2.2 pp, high confidence |
| HLE | 32.5% | estimated ± 9.5 pp, medium confidence |
| IFBench | 77.0% | measured |
| AA-LCR | 83.3% | measured |
| LiveCodeBench v6 | 85.3% | estimated ± 2.1 pp, high confidence |
| AA-Omniscience Accuracy | 27.0% | measured |
| SciCode | 43.6% | measured |
| Terminal-Bench 2.1 | 51.7% | measured |
| BrowseComp | 78.3% | estimated ± 3.1 pp, high confidence |
| HLE w/ tools | 39.0% | estimated ± 6.1 pp, low confidence |
| IFEval | 92.8% | estimated ± 3.5 pp, high confidence |
| LongBench v2 | 61.5% | estimated ± 5.9 pp, low confidence |
| VITA-Bench | 46.4% | estimated ± 9.9 pp, low confidence |
| τ²-bench results | 89.0% | estimated ± 2.4 pp, medium confidence |
| AA Coding Index | 49.0% | measured |
| AA-MMMU-Pro | 74.3% | measured |
| APEX-Agents | 30.3% | estimated ± 0.6 pp, medium confidence |
| APEX-Agents-AA | 34.3% | estimated ± 5.7 pp, medium confidence |
| BioMysteryBench (human-difficult) | 32.8% | estimated ± 5.9 pp, low confidence |
| DeepSearchQA | 74.6% | measured |
| SWE-bench Verified | 76.0% | measured |
| AutomationBench | 48.1% | estimated ± 7.0 pp, low confidence |
| BFCL v4 | 70.6% | estimated ± 2.6 pp, high confidence |
| CyberGym | 53.6% | estimated ± 5.3 pp, low confidence |
| JobBench | 45.9% | estimated ± 6.8 pp, medium confidence |
| SkillsBench | 44.3% | measured |
| SWE-bench Pro | 51.2% | measured |
| τ³-bench results | 69.6% | estimated ± 1.8 pp, low confidence |
| WideResearch | 73.0% | estimated ± 1.5 pp, low confidence |
| AA Agentic Index | 10.5% | measured |
| MMLU-Pro | 85.6% | estimated ± 3.5 pp, high confidence |
| GPQA | 82.5% | estimated ± 3.2 pp, high confidence |
| AA-IFBench | 78.8% | estimated ± 8.5 pp, medium confidence |
| Gert Labs | 52.9% | estimated ± 7.2 pp, medium confidence |
| AA-AnalystAgent | 16.4% | estimated ± 3.6 pp, medium confidence |
| AA EnterpriseOps-Gym | 34.7% | measured |
| AA Harvey LAB | 79.4% | estimated ± 1.2 pp, low confidence |
| ApprenticeBench | 0.0% | estimated ± 5.2 pp, low confidence |
| ARC-AGI-1 | 32.9% | estimated ± 5.4 pp, low confidence |
| ARC-AGI-2 | 10.0% | estimated ± 14.2 pp, low confidence |
| CursorBench 3.1 | 31.4% | estimated ± 3.4 pp, low confidence |
| CursorBench 3.2 | 42.8% | estimated ± 3.6 pp, medium confidence |
| Terminal-Bench 3.0 | 3.8% | estimated ± 9.1 pp, low confidence |
| FrontierCode 1.1 Main | 13.1% | estimated ± 2.7 pp, low confidence |
| FrontierSWE v2 | 0.0% | estimated ± 9.0 pp, low confidence |
| OfficeQA Pro | 50.3% | estimated ± 7.2 pp, low confidence |
| OSWorld-Verified | 65.9% | measured |
| Vals GPQA Diamond | 77.7% | estimated ± 7.4 pp, medium confidence |
| Vals LiveCodeBench | 84.7% | estimated ± 0.8 pp, medium confidence |
| Vals MMLU-Pro | 84.4% | estimated ± 2.2 pp, high confidence |
| Vals SWE-bench | 69.3% | estimated ± 5.3 pp, low confidence |
| Terminal-Bench 2.1 (Vals AI) | 51.1% | estimated ± 6.0 pp, low confidence |
| AA AutomationBench | 6.8% | measured |
| GDP.pdf | 10.0% | measured |
| AA ITBench | 43.3% | estimated ± 3.6 pp, medium confidence |
| AA Tau3 Banking | 23.5% | measured |
| AA Terminal-Bench 2.1 | 51.7% | measured |
| AA Terminal-Bench 4.0 | 0.5% | measured |
| CursorBench 4.0 | 15.5% | estimated ± 1.6 pp, medium confidence |
| CWE-bench v1 | 37.0% | estimated ± 6.0 pp, medium confidence |
| DeepSWE | 12.7% | estimated ± 6.4 pp, low confidence |
| HLE w/o tools | 19.0% | estimated ± 5.7 pp, medium confidence |
| OSWorld 2.0 | 0.0% | estimated ± 3.6 pp, medium confidence |
| PostTrainBench v1.1 | 20.6% | estimated ± 4.7 pp, medium confidence |
| Terminal-Bench 4.0 | 4.5% | estimated ± 5.1 pp, low confidence |
| Terminal-Bench-Science 0.1 | 0.0% | estimated ± 4.1 pp, low confidence |
| Toolathlon-Verified | 73.0% | estimated ± 2.2 pp, low confidence |
| Vibe Code Bench | 6.7% | estimated ± 8.1 pp, medium confidence |
| CharXiv | 78.8% | measured |
| CharXiv w/o tools | 77.9% | estimated ± 1.8 pp, medium confidence |
| ExploitGym | 13.6% | estimated ± 5.0 pp, low confidence |
| C-Eval | 90.0% | estimated ± 1.0 pp, low confidence |
| Claw-Eval | 69.7% | estimated ± 5.7 pp, medium confidence |
| DeepPlanning | 47.9% | estimated ± 13.1 pp, low confidence |
| HMMT Feb 2025 | 94.7% | estimated ± 1.9 pp, medium confidence |
| HMMT Feb 2026 | 85.6% | estimated ± 4.0 pp, high confidence |
| MathVision | 90.4% | estimated ± 2.4 pp, medium confidence |
| MCP Atlas | 75.5% | measured |
| MCP-Tasks | 77.0% | estimated ± 0.6 pp, low confidence |
| MMAnswerBench | 82.3% | estimated ± 1.3 pp, high confidence |
| MMLU-Redux | 95.0% | estimated ± 1.1 pp, medium confidence |
| MMMU-Pro | 74.0% | measured |
| NL2Repo | 35.3% | estimated ± 5.1 pp, medium confidence |
| QwenClawBench | 62.7% | estimated ± 2.6 pp, high confidence |
| ScreenSpot Pro | 75.4% | measured |
| SuperGPQA | 72.0% | estimated ± 7.5 pp, medium confidence |
| Toolathlon | 27.0% | estimated ± 6.3 pp, low confidence |
| VideoMMMU | 84.1% | estimated ± 1.0 pp, high confidence |
| ERQA | 65.3% | estimated ± 3.7 pp, high confidence |
| HealthBench Hard | 3.8% | estimated ± 7.7 pp, low confidence |
| LiveCodeBench Pro | 76.4% | estimated ± 6.0 pp, low confidence |
| MedXpertQA (MM) | 60.8% | estimated ± 2.4 pp, high confidence |
| MMLU-Pro (Arcee) | 86.9% | estimated ± 4.0 pp, medium confidence |
| React Native Evals | 76.4% | estimated ± 3.5 pp, high confidence |
| ResearchClawBench | 18.7% | estimated ± 1.8 pp, medium confidence |
| SWE-Rebench | 57.8% | estimated ± 3.6 pp, medium confidence |
| ARC-AGI-3 | 1.7% | estimated ± 3.4 pp, medium confidence |
| BioMysteryBench (human-solvable) | 81.7% | estimated ± 0.6 pp, low confidence |
| DRACO | 67.5% | estimated ± 3.0 pp, low confidence |
| FrontierCode 1.1 Extended | 52.8% | estimated ± 3.0 pp, low confidence |
| HealthBench (raw) | 30.3% | 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) | 42.2% | estimated ± 5.0 pp, low confidence |
| HLE-Verified | 15.3% | estimated ± 2.0 pp, low confidence |
| LABBench2 | 67.5% | estimated ± 1.8 pp, low confidence |
| AIME 2025 | 95.4% | estimated ± 1.5 pp, medium confidence |
| MMMU | 82.4% | estimated ± 0.9 pp, high confidence |
| LiveCodeBench | 96.2% | estimated ± 8.9 pp, low confidence |
| Agents' Last Exam | 23.0% | estimated ± 6.3 pp, low confidence |
| OpenHarmony Bench | 51.0% | estimated ± 2.4 pp, high confidence |
| SimpleVQA | 82.7% | estimated ± 5.7 pp, low confidence |
| MMLU | 89.8% | estimated ± 1.1 pp, medium confidence |
| MMMU-Pro w/ Python | 75.6% | estimated ± 0.6 pp, high confidence |
| MATH-500 | 97.6% | estimated ± 3.1 pp, medium confidence |
| MathVision w/ Python | 76.8% | estimated ± 0.9 pp, low confidence |
| RealWorldQA | 84.9% | estimated ± 1.1 pp, medium confidence |
| Video-MME (with subtitle) | 87.2% | estimated ± 0.9 pp, medium confidence |
| OmniDocBench 1.5 | 75.8% | measured |