Loading page content
Muse Glimmer 30BOpen
Tool use · Reasoning
Evaluation overview
Overall rank
#66
73.9 points · 194 models
Open-weight rank
#21
75.9 points · 116 models
Data coverage
32 items
2 sources · updated through 2026-08-15
Domain performance
Coding#68 / 174
69.9
Research#67 / 188
82.3
Agents#47 / 140
73.2
Vision#50 / 74
70.3
Long context#4 / 92
97.8
Evaluation highlights
All evaluationsBasic information
- Context (input / output)
- 131K / -
- Parameters
- 29.6B
- Release date
- 2026-08
- License
- Apache 2.0
- Input modalities
- Text · Image
- Output modalities
- Text
- Architecture
- Dense Causal Transformer + Perception Encoder
- Total parameters
- ~29.6B (including perception encoder)
- Attention
- 3 local + 1 global repeating; 2,048-token sliding window
- Default reasoning effort
- high (supports low / medium / high / xhigh)