All tasks / Models
Muse Glimmer 30B
Meta's ~30B-parameter open-weights Muse model under Apache 2.0, small enough for a single consumer GPU at 4-bit.
- Meta
- Open weights
- Released Aug 2026
Where it ranks
- AChat & assistantsModels · Local · 1,425 on LMArena Text Arena (overall, style control)
- BWritingModels · Local · 1,364 on LMArena Text Arena: Creative Writing (style control)
- CCodingModels · Local · 1,355 on LMArena WebDev / Code Arena (overall)
- BImage understandingModels · Local · 74.3% on Artificial Analysis MMMU-Pro (independent evaluation) (secondary benchmark)
Facts
- Vendor
- Meta
- Family
- Muse Glimmer
- Released
- Aug 2026
- Weights
- Open: download and run it yourself
- License
- Apache-2.0
- Parameters
- 29.6B
- GPU memory to run locally
- ~19 GB (4-bit ≈ 29.6 B × 0.57 GB + 1.5 GB ≈ 18–19 GB, estimate; official GGUF repo published.)
- API price
- $0.3 in / $1.2 out per 1M tokens
- Access
- Free (open weights)
- Last verified
- 18 Oct 2026
Price from openrouter.ai, 18 Oct 2026.
Benchmark results
| Benchmark | Result | Entry on the leaderboard | As of | Source |
|---|---|---|---|---|
| Artificial Analysis Intelligence Index | 17 | Muse Glimmer (High)Raw value from the page's embedded model data; the site displays it rounded to an integer. | 18 Oct 2026 | artificialanalysis.ai |
| Artificial Analysis MMMU-Pro (independent evaluation) | 74.3% | Muse Glimmer (High)Raw fraction 0.7434 from the model data embedded in https://artificialanalysis.ai/leaderboards/models, converted to percent; rank 53 of 124 entries with an MMMU-Pro result; no last-updated date shown. | 19 Oct 2026 | artificialanalysis.ai |
| EQ-Bench Creative Writing v3 | 1,798 | meta-models/Muse-Glimmer-30BElo score from the leaderboard data embedded in https://eqbench.com/creative_writing.js (file last modified 2026-09-24); rank 26 of 140; rubric score 16.26. | 24 Sep 2026 | eqbench.com |
| LMArena Text Arena (overall, style control) | 1,425 | muse-glimmerFrom LMArena's official leaderboard-dataset (text_style_control, split latest, category overall, published 2026-10-15), which mirrors https://arena.ai/leaderboard/text (site blocks automated fetches). | 15 Oct 2026 | huggingface.co |
| LMArena Text Arena: Creative Writing (style control) | 1,364 | muse-glimmerFrom LMArena's official leaderboard-dataset (text_style_control, split latest, category creative_writing, published 2026-10-15), which mirrors https://arena.ai/leaderboard/text/creative-writing (site blocks automated fetches). Rank 166 of 414; 736 votes; 95% CI 1341.0-1386.4. | 15 Oct 2026 | huggingface.co |
| LMArena WebDev / Code Arena (overall) | 1,355 | Muse GlimmerFrom LMArena's official leaderboard-dataset (webdev = Code Arena, split latest, category overall, published 2026-10-18), which mirrors https://arena.ai/leaderboard/webdev. Rank 114; 1646 votes; 95% CI 1339.6-1370.9. | 18 Oct 2026 | huggingface.co |
| SWE-bench Verified (vendor-reported) | 76% | Muse Glimmer-30B (High Reasoning)vendor-reported in model card, harness: not stated | 18 Oct 2026 | huggingface.co |
Where you can use it
- SDeepInfraRuntimes & gateways · Cloud
- SHugging Face Inference ProvidersRuntimes & gateways · Cloud
- SHuggingChatApps · Cloud
- ANVIDIA NIM APIs (build.nvidia.com)Runtimes & gateways · Cloud
- SOllamaRuntimes & gateways · Local
- SPoeApps · Cloud
- SRequestyRuntimes & gateways · Cloud
- SSGLangRuntimes & gateways · Local
- STogether AIRuntimes & gateways · Cloud
- SVercel AI GatewayRuntimes & gateways · Cloud
Sources
- https://huggingface.co/meta-models/Muse-Glimmer-30B
- https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset
- https://artificialanalysis.ai/leaderboards/models
- https://arena.ai/leaderboard/text/creative-writing
- https://eqbench.com/creative_writing.html
- https://ollama.com/search?c=vision
- https://artificialanalysis.ai/evaluations/mmmu-pro