MetaMuse
Muse Glimmer 30B
Meta · Muse. Released Aug 10, 2026.
Muse Glimmer 30B is a model from Meta in the Muse family, released Aug 10, 2026. evals.report tracks 13 reported Muse Glimmer 30B benchmark scores across MCP Atlas, SWE-bench Pro, SWE-bench Verified, Terminal-Bench 2.1, SciCode, CharXiv, ScreenSpot-Pro, MMMU-Pro, and 5 more — each shown with its benchmark, metric, source status, and date, and never combined into a single ranking.
Open13 results
Benchmark results 13
Compare this model| Benchmark | Category | Score | Metric | Status | Date | |
|---|---|---|---|---|---|---|
| MCP Atlas | Tool use | 75.5% | pass rate | Verified | Aug 10, 2026 | Details |
| SWE-bench Pro | Coding | 51.2% | % resolved | Verified | Aug 10, 2026 | Details |
| SWE-bench Verified | Coding | 76.0% | % resolved | Verified | Aug 10, 2026 | Details |
| Terminal-Bench 2.1 | Agents | 51.7% | task success | Verified | Aug 10, 2026 | Details |
| SciCode | Coding | 43.6% | accuracy | Verified | Aug 10, 2026 | Details |
| CharXiv | Multimodal | 78.8% | accuracy | Verified | Aug 10, 2026 | Details |
| ScreenSpot-Pro | Multimodal | 75.4% | accuracy | Verified | Aug 10, 2026 | Details |
| MMMU-Pro | Multimodal | 74% | accuracy | Verified | Aug 10, 2026 | Details |
| OSWorld | Agents | 65.9% | task success rate | Verified | Aug 10, 2026 | Details |
| IFBench | Reasoning | 77.0% | accuracy | Verified | Aug 10, 2026 | Details |
| GPQA Diamond | Reasoning | 83.5% | accuracy | Verified | Aug 10, 2026 | Details |
| Humanity's Last Exam | Reasoning | 22.0% | accuracy | Verified | Aug 10, 2026 | Details |
| AIME 2026 | Reasoning | 94.7% | accuracy | Verified | Aug 10, 2026 | Details |