Muse Glimmer 30B
Meta · released Aug 9, 2026 · meta-models/Muse-Glimmer-30B
- Type
- Open weightsApache License 2.0
- Params
- 29.8B
- Context
- 131K
about 98K words of context
Our take
Written Sep 10, 2026Muse Glimmer 30B is a 29.8-billion-parameter text-and-image model from Meta with a permissive Apache licence. It scores broadly across chat arenas with coding as its relative strength, and sits in the budget tier for hosted inference.
Pick this for general text-and-image tasks where a permissive licence matters — Apache 2.0 allows commercial use, fine-tuning and redistribution. Use it for coding assistance, where it scores highest in its own profile, or for projects needing 131,072 tokens of context with open weights. Skip it if creative writing or web development are central, or if you need the fastest throughput at the cheapest price.
The case for it
- Coding is the relative high point in its arena profile, 54.8 points above its own general text score.
- Truly permissive Apache 2.0 licence allows commercial use, fine-tuning and redistribution.
- Lowest hosted price is matched by three providers, so you are not locked to one host for the cheapest tier.
The case against it
- Creative writing and web development lag its own coding score by 108.8 and 122.6 points respectively.
- No measured speed advantage at the cheapest price point; the fastest host costs more.
- Active parameter count is undisclosed, so efficiency claims are unverified.
How good is it?
EverydayGeneral questions and everyday reasoning
Arena Text (overall)75th of 168 · 1424
CodingWriting and fixing code on its own
Arena Coding69th of 168 · 1480
AgenticPlanning, calling tools, staying on task
Not yet scored on Arena Agent.
WritingDrafting and rewriting prose
Arena Creative Writing92nd of 168 · 1365
Arena Creative Writing is the only board that has scored it for this.
These tests check whether a model follows instructions — a precondition for all the work above, but not a measure of how well that work is done.
Every published score for this model7 scoresEvery figure we hold, from 7 boards, with who ran it and a link to the source — including the boards no rating above is built on.
Can you run it yourself?
Comfortable fit
GeForce RTX 4090 · 24 GB
Room to spare. 2.1 GB spare means a 10% error in the size would not change the answer.
GeForce RTX 5090 · 32 GB
Room to spare. 10.1 GB spare means a 10% error in the size would not change the answer.
Apple M1 Pro (16-core GPU) · 32 GB
Room to spare. 3.3 GB spare means a 10% error in the size would not change the answer.
Memory use by level
Against a 24 GB card.
What is quantisation? →This model on every device we track71 devicesThe Q4 build most people download, on each device: what the weights come to, how much context the memory leaves, and whether it runs. Smallest device that runs it first.
Check against your own machine → · Where to rent it hosted →
Or rent it from someone else
Prices checked 4 hours ago — each listing carries its own date.
- per 1M tokens
- $0.30 in / $1.10 out
- Context served
- 131K
- Throughput
- ~62 tok/s
| Provider | In / out per 1M tokens | Context | Throughput | Trains on prompts | Logs prompts | Zero retention |
|---|---|---|---|---|---|---|
| PhalaThrough OpenRouter | $0.30 / $1.10checked 4 hours ago | 131K118K max reply | 62 tok/s | No | No | Confirmed |
| DeepInfrabf16Direct and through OpenRouter | $0.30 / $1.20checked 4 hours ago | 131K16K max reply through OpenRouter | 94 tok/sthrough OpenRouter | DirectUnknownThrough OpenRouterNo | DirectUnknownThrough OpenRouterNo | DirectUnknownThrough OpenRouterConfirmed |
| OpenRouterOpenRouter's own listing | $0.35 / $1.50checked 4 hours ago | 131K | not measured | Unknown | Unknown | Unknown |
| Together AIThrough OpenRouter | $0.35 / $1.50checked 4 hours ago | 131K118K max reply | 65 tok/s | No | No | Confirmed |
DeepInfra
OpenRouter
Together AIAcross the 4 listings we hold: 3 say they do not train on prompts (1 of them only through OpenRouter), 0 say they do and 1 does not say. 3 appear in the zero-retention registry we check (1 of them only through OpenRouter); the rest are unknown to us.
What each host's API supports
From the parameter list each endpoint publishes. Streaming is omitted: nothing we hold reports it, for any model.
| Provider | Tool calling | JSON output | Strict schema |
|---|---|---|---|
| PhalaThrough OpenRouter | ✓ | ✓ | ✓ |
| DeepInfrabf16Direct and through OpenRouter | ✓ | ✓ | ✓ |
| OpenRouterOpenRouter's own listing | ✓ | ✓ | ✓ |
| Together AIThrough OpenRouter | ✓ | ✓ | ✓ |
Tool calling: 4 of 4 listings say yes. JSON output: 4 of 4 listings say yes. Strict schema: 4 of 4 listings say yes.
When we formed this view
Recent changes
What moved
output −8% ($1.20 → $1.10 per 1M tokens)What moved
input −14% ($0.35 → $0.30 per 1M tokens), output −20% ($1.50 → $1.20 per 1M tokens)What moved
first indexed by our pipelineEach date is the day we first saw the change, or the day the maker announced it.
What we do not know about this model yet
- We hold no measured file for it, so all 3 sizes on this page are calculated from the parameter count.
- Nothing we hold says whether an endpoint streams, so we do not show it either way.
- 1 of 4 listings does not say whether it trains on prompts, and 1 answers only through OpenRouter, not for its own listing.
- We hold no batch or off-peak rate for any of its listings.
- We hold a decode speed for it, but no prompt-processing (prefill) figure, so how long the input side of a job takes is unknown to us.
Licence and identifiers
What the licence allowsApache License 2.0, what it allows commercially, and the identifiers you need to pull this model — its Hugging Face repo, our slug and a machine-readable card.
Licence
Apache License 2.0
Fully permissive: commercial use, redistribution, and derivatives allowed. Requires attribution and a copy of the license. Includes an express patent grant.
Identifiers
- Hugging Face
- meta-models/Muse-Glimmer-30B
- Architecture
- Dense
- Takes in, gives back
- Text and images in, text out
- Catalogue slug
- meta-muse-glimmer-30b