LLM Stats (PinchBench - Meta Muse Glimmer 24 Tasks): leaderboard
Mirror of Meta Muse Glimmer report Appendix Table 5/A.2: 24 scientific Python container tasks, four-run mean reward; fused into the primary report board.
Metric: Mean Task Reward (%). Source: llm-stats.com. 1 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Muse Glimmer 30B (High) | 90.5 | #179 (Muse Glimmer 30B) |
Interactive version: theaggregate.ai/benchmark?slug=llm-stats-pinchbench-meta-muse-glimmer-24-tasks · How It Works · Data refreshed daily, snapshot 2026-10-22.