Meta
Muse Glimmer
Meta's 30B dense model for local agents, released August 10, 2026 with downloadable Apache 2.0 weights. Accepts text and images and produces text, with a 128K-token context window. Quantized language-model weights fit under 20GB; Meta targets 24GB or 32GB total memory for the weights, working memory, perception encoder and speculative-decoding drafter. Supports tool use, coding and multi-step agent workflows on consumer hardware.
- Provider
- Meta
- Context
- 128KCatalog · As of Aug 13, 2026Within 30d
- yno subscription tier
- Pro
- Released
- Aug 10, 2026
- Speed
- Fast
- Reasoning
- Advanced
- Modality
- Multimodal
Benchmarks
Quality
- AA Intelligence Index
- AA Coding Index
- GPQA Diamond
- 83.5%Artificial Analysis · Date unknownAge unknown
Speed and latency
- Output speed
- 102.55tokens/sConfiguration: Prompt length: 1,000 · Parallel queries: 1Artificial Analysis · Date unknownAge unknown
- Time to first token
- 0.44sConfiguration: Prompt length: 1,000 · Parallel queries: 1Artificial Analysis · Date unknownAge unknown
- Time to first answer token
- 19.94sConfiguration: Prompt length: 1,000 · Parallel queries: 1Artificial Analysis · Date unknownAge unknown
API pricing
- API input cost
- API output cost
Benchmark variant: Muse Glimmer (high)
AA data retrieved Sep 11, 2026 · Artificial Analysis
Indexes use points; evaluations use accuracy percentages. API measurements do not measure yno application performance. Methodology
Capabilities
- Open Source
- Agentic
- Local Deployment
- Code
- Fine-tuning
- Vision
Use cases
- On-prem/local agents
- Single-GPU deployment
- Custom fine-tuning
- Data-sovereign workloads
Strengths
- Apache 2.0 weights
- Text and image input
- Local agent workflows
- 24GB/32GB quantized targets
Best for
Teams building local agents with downloadable weights and image understanding
No credit card required
Related models
Frequently asked
- What is Muse Glimmer?
- Meta's 30B dense model for local agents, released August 10, 2026 with downloadable Apache 2.0 weights. Accepts text and images and produces text, with a 128K-token context window. Quantized language-model weights fit under 20GB; Meta targets 24GB or 32GB total memory for the weights, working memory, perception encoder and speculative-decoding drafter. Supports tool use, coding and multi-step agent workflows on consumer hardware.
- How much does Muse Glimmer cost in yno.ai?
- Muse Glimmer is available on the Pro tier of yno.ai.
- What can Muse Glimmer do?
- Muse Glimmer is best for Teams building local agents with downloadable weights and image understanding. Its main capabilities include Open Source, Agentic, Local Deployment, Code, Fine-tuning, Vision.
- How does Muse Glimmer compare to other models?
- Muse Glimmer excels at Apache 2.0 weights and is recommended for On-prem/local agents, Single-GPU deployment, Custom fine-tuning, Data-sovereign workloads. See related models below.
- How do I use Muse Glimmer in yno.ai?
- Sign up for yno.ai, select Muse Glimmer from the model picker, and start chatting.