Skip to content
ModelCap

Muse Glimmer 30B

Official publisher releaseOpen weights6.9d #57 → #56

Meta·#56 of 371 on the current board·cheaper than #52 Qwen3.5 397B A17B

Muse Glimmer 30B is a language model from Meta. It ranks #56 of 371 on ModelCap's current board, with an Index score of 66.8, modeled from the publisher's launch results against measured models. Its score range of 51.2–82.5 spans positions #7–#130. It is listed at $0.35 per million input tokens and $1.50 per million output tokens across 3 providers. It reads up to 131K tokens of context, and its weights are openly downloadable.

Figures as of 19 Nov 2026, 05:16 UTC · download the public dataset · methodology

ModelCap Index · Current board
66.8#56
Modeled from the publisher's launch results against measured models · range 51.2–82.5
Market Gravity
38.8Increased by 0.4
Secondary market signal
Input
$0.35
per 1M tokens
Output
$1.50
per 1M tokens
Context
131K
tokens
Providers
3
endpoints

Providers

Uptime measured over the last 30 minutes
ProviderInputOutputContextQuantUptime
DeepInfrabf16Cheapest$0.30$1.20131Kbf16100.0%
ModelRunbf16$0.30$1.20131Kbf16100.0%
Together$0.35$1.50131K—100.0%

Hugging Face Inference Providers

Exact repository identity

Provider economics and telemetry published by the Hugging Face Router. These are operational signals, not benchmark results.

ProviderInputOutputContextTTFTThroughputFeatures
deepinfra
live
$0.30$1.20131K690 ms131.0 tok/stools · structured
modelrun
live
$0.30$1.20131K507 ms79.0 tok/stools · structured
together
live
$0.35$1.50131K1053 ms35.0 tok/stools · structured
featherless-ai
live
——————

Reported evaluation leads (115)

Not independent measurement

Structured rows reported in this exact model repository at the pinned revision below. They are the publisher talking about itself: never measured evidence, never a score input on their own. Rows marked as used were laddered against independently measured peers inside the launch-card channel of this model's modeled placement, with the replay-measured optimism haircut applied; the others played no part.

Dataset / taskMetricValueRevisionProvenance
allenai/IFBench_test
prompt_level_loose
—77
Unpinned
not published
Reported
Not used in launch estimate
benchflow/skillsbench
skillsbench_v1_1
—44.3
Unpinned
not published
Reported
Not used in launch estimate
cais/hle
hle
—22
Unpinned
not published
Reported
Not used in launch estimate
facebook/mmi
mmi-value
—11.58
Unpinned
not published
Reported
Not used in launch estimate
facebook/mmi
modality-presence-score
—15.54
Unpinned
not published
Reported
Not used in launch estimate
facebook/mmi
error-rate
—41.1
Unpinned
not published
Reported
Not used in launch estimate
facebook/mmi
pass-rate-strict
—9.18
Unpinned
not published
Reported
Not used in launch estimate
facebook/mmi
pass-rate-lenient
—10.86
Unpinned
not published
Reported
Not used in launch estimate
facebook/mmi
recall
—14.28
Unpinned
not published
Reported
Not used in launch estimate
facebook/mmi
precision
—99.78
Unpinned
not published
Reported
Not used in launch estimate
facebook/multiloko
average_em
—25.87
Unpinned
not published
Reported
Not used in launch estimate
facebook/multiloko
gap_em
—40.8
Unpinned
not published
Reported
Not used in launch estimate
facebook/multiloko
average_contains
—31.85
Unpinned
not published
Reported
Not used in launch estimate
facebook/multiloko
gap_contains
—35.6
Unpinned
not published
Reported
Not used in launch estimate
facebook/multiloko
em_arabic
—22.4
Unpinned
not published
Reported
Not used in launch estimate
facebook/multiloko
em_bengali
—27.2
Unpinned
not published
Reported
Not used in launch estimate
facebook/multiloko
em_cantonese
—18.4
Unpinned
not published
Reported
Not used in launch estimate
facebook/multiloko
em_czech
—26.4
Unpinned
not published
Reported
Not used in launch estimate
facebook/multiloko
em_dutch
—24.4
Unpinned
not published
Reported
Not used in launch estimate
facebook/multiloko
em_english
—53.6
Unpinned
not published
Reported
Not used in launch estimate

Showing 20 of 115 rows.

ModelCap Index · Current board

Methodology
66.8
Modeled from the publisher's launch results against measured models
Current board
#56 of 371
Score range
51.2–82.5
Rank range
#7–#130
Technical details
Public rank basis
Modeled · peer benchmarks (architecture)
Evidence label
Modeled · peer benchmarks · 14 reported rows against measured corpus ladders (71%); publisher-corpus-prior over 214 held-out anchors (29%); shrunk 5.9 up toward the measured corpus
Index support
40.0% · limited
Identity binding
Exact catalogue product · meta/meta/muse-glimmer-30b · aggregates disclosed benchmark configurations · endpoint configuration not claimed · artifact metadata meta-models/Muse-Glimmer-30B@1297fd7dfeac4598948a5d15886968435c81c2b7 (not the evaluation revision)
Index observations used
0
Qualified Index evidence point
66.8
How this position was produced · fused launch placement

No general-family board lists this model yet. Its point estimate is the precision-weighted mean of every channel that spoke for it; each channel’s share is the weight it carried. Independent measurements replace this placement as boards list the model.

Fused estimate
66.8 ± 12.3
Independent share
0%
Shrink toward corpus
+5.9
Optimism haircut
None applied
Method
precision-weighted-v1
Channels fused into this model’s launch placement
ChannelEstimateShare
Corpus ladder · not independent
14 reported rows against measured corpus ladders
60.9 ± 14.571%
Publisher corpus prior · not independent
publisher-corpus-prior over 214 held-out anchors
81.7 ± 22.929%
Bayesian Evidence Score · measured evidence input only

BES normalizes admitted public benchmark evidence for measured Index rows. Its legacy score and observables prior do not define the public language rank.

Observed capability
—
Evidence status
No admitted measured evidence
General preference
—
Coding
—
Agents & tools
—
Reasoning
—
Evidence breadth
0%
Evidence coverage
0%
Market and catalogue signals · secondary, never the language Index rank
Usage (OpenRouter popularity)
19.4
Liquidity (providers × uptime)
64.3
Open reach (HF downloads)
57.9
Surface (context / tools / modalities)
85.2
Freshness
68.0
  • · OpenRouter popularity #111
  • · 3 live providers
  • · HF 30d downloads 155703
  • · Catalogue: context 131072, reasoning, tools, modalities Text/Image
  • · First seen 2026-08-09

These adoption and deployment observations remain context only. They do not change this model's ModelCap Index score or rank.

The bounded 0–100 capability score blends each source’s competitive placement with its published achievement, then combines capability families. Evidence breadth adds a modest uncertainty adjustment; price and popularity are not part of benchmark-led rank.

Experimental capability estimate

Estimate policy
47.9
34.3–61.4 uncertainty interval
Excluded from rank
Confidence
79%
Training support
225 models · 132 lineages
Out-of-domain check
in domain
Feature coverage
100%
Lineage-held-out error
8.1 MAE

Experimental estimate. This layer predicts from admitted evidence and safe metadata only. It never enters ModelCap Score or rank.

Deployment readiness

Metadata index
74.4
partial · 76% metadata coverage
Not capability
Artifact reproducibility
85.0
Access & legal clarity
100.0
Deployability
96.3
Evaluation provenance
0.0

Not yet published: the base model it derives from, admitted benchmark results, results across several capability areas, results from more than one source, a confident match to its benchmark entries, pinned evaluation dataset versions.

A metadata completeness and deployability index—not a safety certification, quality grade, or production-readiness claim.

Market activity

Venue coverage
Market Gravity
38.8
OpenRouter weekly popularity
#111 of 345

Market Gravity uses OpenRouter’s full-catalogue popularity order. Sparse Vercel top-ten observations appear only when published and do not affect the score.

Arena preference

No matched Arena rating

This model is visible because it is available in the live catalogue, but it has no trustworthy community result yet. ModelCap does not infer a quality score from price, features, or market activity.

Market Gravity breakdown

Usage55%
19.4

OpenRouter weekly popularity across the full model catalogue.

Liquidity25%
64.3

Independent providers versus the model's open or closed cohort, weighted by uptime.

Open reach15%
57.9

Hugging Face 30-day downloads for open models; neutral for closed models.

Freshness5%
68.0

Time since first public availability, on a six-month half-life.

Specification

Model ID
meta/muse-glimmer-30b
Context window
131K tokens
Declared output cap
118K tokens

OpenRouter catalogue or provider-endpoint declaration. Limits can vary by provider and prompt length; a model-wide maximum has not been independently verified. OpenRouter metadata

Inputs
Text, Image
Outputs
Text
Tokenizer
Other
Cached input tokens
$0.04 / 1M
First seen on OpenRouter
9 Aug 2026

Capabilities

  • Supported: Reasoning
  • Supported: Tool use
  • Supported: Structured output
  • Supported: Response format
  • Not supported: Moderated

Weights & access

Open weights. Downloadable weights are published under an OSI or free-culture license in the current repository metadata. This describes weight availability, not whether the full training stack qualifies as Open Source AI.

Access
Open weights
Licence
apache-2.0
Pinned revision
1297fd7dfeac
Parameters
29.8B
Architecture
MuseGlimmerForConditionalGeneration
Model type
muse_glimmer
Hugging Face downloads (30d)
156K
Downloads (all time)
1.2M
Likes
2K
Repository updated
18 Nov 2026, 23:58 UTC

Benchmark leaderboards featuring Muse Glimmer 30B

All leaderboards
Public benchmark boards on which Muse Glimmer 30B has a published result
LeaderboardScoreSource rankConfigurationPublished
WildClawBench OpenClaw leaderboardWildClawBench47.6#16 of 34Default20 Jul 2026

Alternatives to Muse Glimmer 30B, by the numbers

Ranked neighbours

Models within 3 places of #56 on the ModelCap Index.

Cheaper at a similar or higher Index

Index score within 3 points of, or above, this model's, with a lower listed blended price (3:1 input:output).

Other ranked Meta models

The same publisher's other current models with a public Index position.

Muse Glimmer 30B: common questions

How does Muse Glimmer 30B rank among AI models?

Muse Glimmer 30B holds ModelCap Index position #56 of 371 ranked language models as of 19 November 2026, with an Index score of 66.8 (interval 51.2–82.5); evidence: modeled · peer benchmarks. The Index combines public benchmark boards with published uncertainty; the methodology page explains the weighting.

How much does Muse Glimmer 30B cost per 1M tokens?

Muse Glimmer 30B is listed at $0.35 per 1M input tokens and $1.50 per 1M output tokens as of 19 November 2026; the lowest output price among 3 providers is $1.20 on ModelRun. Prices come from the OpenRouter catalogue and re-observe on every refresh. For 1,000 requests using 2,000 input and 500 output tokens each, the listed rates imply $1.45 total (2M input + 0.5M output). This excludes caching, batch discounts, prompt-length tiers, tool charges and retries; verify the chosen endpoint's rate and limits before budgeting.

What is the context window of Muse Glimmer 30B?

Muse Glimmer 30B has a published context window of 131K tokens (131,072). The snapshot lists a 117,964 token catalogue-declared route output cap. OpenRouter catalogue or provider-endpoint declaration. Limits can vary by provider and prompt length; a model-wide maximum has not been independently verified. The providers table lists each endpoint's context limit.

Is Muse Glimmer 30B open-weight?

Open weights (apache-2.0). Downloadable weights are published under an OSI or free-culture license in the current repository metadata. This describes weight availability, not whether the full training stack qualifies as Open Source AI. The repository is meta-models/Muse-Glimmer-30B on Hugging Face.

Which API providers serve Muse Glimmer 30B?

3 providers serve Muse Glimmer 30B through OpenRouter as of 19 November 2026: ModelRun, DeepInfra and Together. Each provider's price, context limit, quantization and measured uptime are in the providers table above.

Which benchmarks has Muse Glimmer 30B been evaluated on?

Muse Glimmer 30B has published results on 1 public board tracked by ModelCap as of 19 November 2026: WildClawBench OpenClaw 47.6 (#16 of 34). Each board page ranks every tracked model on that benchmark. The ModelCap Index scores the results its methodology admits, with published uncertainty, and shows reference boards without scoring them.

When was Muse Glimmer 30B released?

Muse Glimmer 30B first appeared in the catalogue on 9 Aug 2026. It is the current version in its family. Rank and price movements since then are recorded on the site's changes feed.