Muse Glimmer
Meta · US · #12 in our ranking · Best for local agents
Muse Glimmer is an open-weight dense model from Meta (United States), released Aug 2026, with 30B parameters, a 128K-token context window, and the Apache 2.0.
Meta built Muse Glimmer for tool use, multi-step reasoning, and recovering from failed tool calls on consumer hardware. Quantized to about 4 bits, the language model fits under 20GB.
Muse Glimmer specs
- Provider
- Meta
- Origin
- US
- Architecture
- Dense
- Parameters
- 30B total
- Context
- 128K tokens | 98K words
- License
- Apache 2.0 Commercial use
- Input
- Text, Image
- Released
- Aug 2026
- Weights
- Hugging Face
Sourced from the model card on Hugging Face. Last checked 2026-10-16.
Highlights
- Apache 2.0 license
- Fits a 24GB GPU when quantized
- Image input
Where it fits for payment teams
A good small model for routine steps that must stay on your own hardware, like tagging tickets or drafting notes.
Harnesses in Runtime
Run Muse Glimmer through any of these harnesses on Runtime.
Provider-reported benchmarks
| MCP Atlas | 75.5 |
|---|---|
| SWE-bench Verified | 76.0 |
| τ3-Banking | 23.5 |
Benchmark scores are reported by each model provider on its Hugging Face model card. They are not independently verified by Runtime.
How to run Muse Glimmer
Download the weights
Pull meta-models/Muse-Glimmer-30B from Hugging Face. Check the Apache 2.0.
Serve it in your cloud
vLLM or SGLang on your GPUs, or a managed host.
Put it to work in Runtime
Point an agent or skill at it, then compare with evals.
Muse Glimmer FAQ
What is the context window of Muse Glimmer?
+
Muse Glimmer supports a 128K-token context window (131,072 tokens), roughly 98K words of English text.
How many parameters does Muse Glimmer have?
+
Muse Glimmer is a dense model with 30B parameters, all active on every token.
Is Muse Glimmer free for commercial use?
+
Yes. Muse Glimmer is released under the Apache 2.0 license, which is permissive and allows commercial use and fine-tuning.
What can Muse Glimmer take as input?
+
Muse Glimmer accepts text, image input and generates text.
Can I run Muse Glimmer in my own cloud with Runtime?
+
Yes. Download the weights from Hugging Face (meta-models/Muse-Glimmer-30B) and serve them with vLLM or SGLang in your cloud, or use a managed provider that hosts the model. Runtime agents can then use Muse Glimmer through a harness like OpenCode, with your data staying in your environment.