$ cat wiki/models/muse-glimmer.md
Muse Glimmer
Compared with
- Qwen 3.8 27B
- DeepSeek V4-Pro-0813
- Gemini 3.7 Flash
- Grok 4.6
- Claude Opus 5
- Laguna S 2.1
- Kimi K3
- Inkling
- GPT-5.6 Sol
- LongCat-2.0
- GLM-5.2
- MiniMax M3
- Muse Spark 1.2
- Qwen 3.8 Max
- DeepSeek V4-Flash
- Gemini 3.5 Flash-Lite
- Gemini 3.6 Flash
- DeepSeek V4
- Grok 4.5
- Claude Sonnet 5
- Claude Fable 5
- Claude Opus 4.8
- Gemini 3.5 Flash
- Grok Build
- Claude Opus 4.7
Spec
| Attribute | Value |
|---|---|
| Developer | Meta / Meta Superintelligence Labs (MSL) |
| Released | 2026-08-10 |
| Announced | 2026-08-10 |
| Context window | 131,072 |
| Pricing | unknown |
| License | Apache 2.0 |
| Availability | Hugging Face (open weights), Ollama, LM Studio, vLLM, SGLang, Together AI, Fireworks AI, OpenRouter |
| 30B dense multimodal, built to run offline on consumer hardware. 4-bit | |
| quantization is reported to cut the memory requirement from 55 GB to 18–20 GB, | |
| which puts it on a single consumer GPU — one write-up runs it on 24 GB of VRAM | |
| (source). Vocabulary | |
| 202,048 tokens, knowledge cutoff 2026-01-04, 100+ languages, and | |
| compatibility with agent frameworks including OpenClaw | |
| (source). |
Pricing is unknown rather than free: the weights are Apache 2.0, and no
first-party hosted API rate was published in anything read. Third-party serving
exists (Together AI, Fireworks AI, OpenRouter) but at those vendors' prices, not
Meta's.
The Context window row records the 131,072 that the model card is reported to
carry; the coverage phrases it as "131,072+", and no source read states what the
"+" refers to (source).
Release Date
2026-08-10, with weights on Hugging Face the same day (source).
Benchmarks
Meta's own figures as relayed by secondary coverage, against the two open models in its size class:
| Benchmark | Muse Glimmer | Gemma4-31B | Qwen3.6-27B |
|---|---|---|---|
| MCP Atlas | 75.5 | 54.2 | 62.5 |
| DeepSearch QA | 74.6 | unknown | unknown |
| Gaia2 | 43.3 | unknown | unknown |
| SWE-Bench Pro | 51.2 | unknown | unknown |
| τ-Bench is named among the suites Meta reports, with no figure surfacing in | |||
| anything read (source). |
Two limits on these numbers, both worth stating before the comparison is used.
No first-party benchmark table was read — the sandbox could not reach Meta's
announcement or the model card. And this repo's sources/evals/ snapshots carry
no column for MCP Atlas, DeepSearch QA or Gaia2, so there is nothing local to
check them against, the same gap recorded on Grok Imagine Image 2.0.
The benchmark selection is itself the claim: every suite named measures completing a task inside a scaffold rather than answering a question. That is consistent with a model distilled for agent work, and it means the figures say little about how Glimmer compares on the knowledge and reasoning benchmarks the frontier models are ranked by.
Use Cases
- Local coding agents and other always-on agent workflows — the stated design target, and the reason the memory figure matters more than the parameter count (source).
- LLM-as-a-judge evaluation, named by Meta as a use the small size suits (source).
- Long-horizon reasoning, tool calling, failure recovery, coding and multimodal reasoning are the capability areas Meta lists for agentic workloads (source).
Distribution is unusually broad for a launch day: Ollama, LM Studio, vLLM,
SGLang, Together AI, Fireworks AI and OpenRouter at release, with llama.cpp,
MLX and ExecuTorch described as landing "in the coming days"
(source). A community GGUF
quantization (unsloth/Muse-Glimmer-30B-GGUF) appeared the same day.
Compared To
| Model | Weights | License | Params | Runs locally |
|---|---|---|---|---|
| Muse Glimmer | released | Apache 2.0 | 30B dense | single consumer GPU (18–20 GB at 4-bit) |
| Muse Spark 1.2 | not released | unknown | undisclosed | no — API only |
| LFM2.5-2.6B | released | not yet read | 2.69B | phone, under 2.5 GB |
| Inkling | released | Apache 2.0 | 975B MoE / 41B active | no |
| The interesting comparison is the first row against the second. Glimmer is | ||||
| distilled from Muse Spark (1.0 / 1.1) — pre-training used **logit distillation | ||||
| on Muse Spark's outputs**, with mid-training adding longer-context, agent-heavy | ||||
| data and richer reasoning traces | ||||
| (source). So Meta is | ||||
| shipping the teacher as a closed paid API and the student as Apache 2.0, | ||||
| on the same product line, five days apart. That is a deliberate split rather than | ||||
| a change of position, and it is the shape Open-Weights Policy Fight has | ||||
| been tracking. |
Conflicting Reports
None. The licence was the one figure worth checking — an early summary did not confirm Apache 2.0, and VentureBeat, Open Source For You and Phoronix each name it explicitly (source).
Open Questions
- Training compute and data — undisclosed. The distillation lineage is stated; the token budget is not.
- What the "+" in "131,072+" means. No source read resolves it.
- Is there a first-party hosted endpoint, and at what price?
- The comparison set is Gemma4-31B and Qwen3.6-27B, both chosen by Meta. How does Glimmer measure against DeepSeek V4-Flash or Kimi K3, the open models Zuckerberg's own framing names as the competition?
- Apache 2.0 covers the weights. No source read states whether the evaluation harness that produced the benchmark figures was released with them — the distinction Eval Harness Configuration exists to record.
Sources
- Meta AI Research — Introducing Muse Glimmer: An Open Agentic Model That Runs on Your Device (source)
- Hugging Face — meta-models/Muse-Glimmer-30B
- Hugging Face Blog — Meta is back with Muse Glimmer
- VentureBeat — Meta returns to open source with Muse Glimmer
- CNBC — Meta launches Muse Glimmer open-weight AI model
- Phoronix — Meta Publishes Muse Glimmer As 30B Open Agentic Model
- MarkTechPost — Meta AI Releases Muse Glimmer