# Muse Glimmer

Meta, released 2026-08-10

**Verdict: conditional.** Muse Glimmer is Meta's 30B open-weight agentic model, distilled from the closed Muse Spark frontier model and released under Apache 2.0. With 4-bit quantization it runs multi-step tool calling, multimodal reasoning, and failure recovery on a single 24GB consumer GPU without cloud dependencies. It leads its size class on agentic orchestration but trails Qwen3.6-27B on terminal and OS automation — reach for it if you need a capable local agent that stays offline, not if you need independent reproducibility or low-level OS control. Last verified 2026-10-13.

## Pricing

- Input: Free (open weights)
- Output: Free (open weights)
- Context: 131K tokens

## Strengths

- Local agentic workflows — runs on a single 24GB consumer GPU at ~20GB quantized, no cloud or internet required
- Tool use and function calling — reliable schema-based tool invocation across long multi-turn agent sessions
- Multimodal reasoning — includes a 1.8B ViT-G/14 perception encoder for screenshots, charts, and documents
- Terminal/OS automation — trails Qwen3.6-27B on TerminalBench 2.1 (51.7 vs 60.7) and OSWorld-Verified (65.9 vs 75.6)
- Reproducibility-critical use — Apache 2.0 weights only; no training data or code released, so the model cannot be independently reproduced

## Benchmarks

- MCP Atlas (Public): 75.5 (https://huggingface.co/meta-models/Muse-Glimmer-30B)
- DeepSearch QA: 74.6 (https://huggingface.co/meta-models/Muse-Glimmer-30B)
- SWE-Bench Pro: 51.2 (https://huggingface.co/meta-models/Muse-Glimmer-30B)
- SWE-Bench Verified: 76.0 (https://huggingface.co/meta-models/Muse-Glimmer-30B)
- TerminalBench 2.1: 51.7 (https://huggingface.co/meta-models/Muse-Glimmer-30B)
- AIME 2026: 94.7 (https://huggingface.co/meta-models/Muse-Glimmer-30B)
- GPQA Diamond (AA): 83.5 (https://huggingface.co/meta-models/Muse-Glimmer-30B)

## Sources

- [TechTimes — Meta Launches Muse Glimmer](https://www.techtimes.com/articles/323787/20260810/meta-launches-muse-glimmer-first-consumer-gpu-agent-model-built-autonomous-tasks.htm)
- [NVIDIA Developer — Run Local Agentic AI Workflows with Muse Glimmer](https://developer.nvidia.com/blog/run-local-agentic-ai-workflows-with-metas-muse-glimmer-on-nvidia/)
- [Hugging Face - Muse Glimmer 30B model card (released, downloadable)](https://huggingface.co/meta-models/Muse-Glimmer-30B)
- [VentureBeat — Meta returns to open source with Muse Glimmer](https://venturebeat.com/technology/meta-returns-to-open-source-with-muse-glimmer-an-apache-2-0-licensed-30b-parameter-ai-model-optimized-for-agents-available-now)
- [MarkTechPost — Meta AI Releases Muse Glimmer](https://www.marktechpost.com/2026/08/10/meta-ai-releases-muse-glimmer/)

Canonical: https://neomanex.com/models/muse-glimmer