Skip to main content

Muse Glimmer-30B

MetaOpen WeightApache 2.0 · Commercial OK

Description

Muse Glimmer-30B is Meta Superintelligence Labs' open-weight, dense multimodal model for autonomous agentic work on consumer hardware. Its 29.6-billion-parameter causal transformer includes a roughly 1.8B-parameter ViT-G/14 perception encoder, accepts interleaved text and images, produces text, and supports a 131,072-token context window. It combines controllable multi-step reasoning, schema-based tool use, coding, failure recovery, and compatibility with agent scaffolds such as OpenClaw and Hermes Agent. Meta also releases 4-bit variants sized for 24 GB and 32 GB systems and a DFlash speculative-decoding drafter. The model is trained on more than 100 languages and released under Apache 2.0.

Release Date
2026-08-10
Parameters
29.6B
Context Length
131K
Modalities
image, text

Capability Radar

60
general
60
coding
50
reasoning
51
scienceest.
50
agents
70
multimodal

Science is estimated from LLM Stats science scores or reasoning when no dedicated science benchmarks are available.

Rankings

Domain#RankScoreSource
Agentic Capability132
33.0
LS
Multimodal Ranking52
45.0
LS

Benchmark Scores (LLM Stats)

(LLM Stats (zeroeval))

Agents

Siren AgentDojo Utility94.2%SR
MCP Atlas75.5%SR
DeepSearchQA74.6%SR
OSWorld-Verified65.9%SR
CI Memories Coverage64.8%SR
Terminal-Bench 2.151.7%SR
SWE-Bench ProPrinceton NLP (2024)51.2%SR
WildClawBench47.6%SR
SkillsBench44.3%SR
GAIA243.3%SR
Siren AgentDojo Attack Success Rate28.4%SR
CI Memories Violation Rate26.4%SR
Tau3 Banking23.5%SR

Biology

GPQANYU + Cohere + Anthropic (2023)83.5%SR
SciCode43.6%SR

Code

SWE-Bench Verified76.0%SR

General

IFBench77.0%SR
MMMU-Pro74.0%SR

Grounding

ScreenSpot Pro75.4%SR

Long Context

AA-LCR80.0%SR
Beam 128K65.1%SR

Math

AIME 202694.7%SR
Humanity's Last Exam22.0%SR

Multimodal

CharXiv-R78.8%SR
OmniDocBench 1.575.8%SR

AA Evaluation Indices

(Artificial Analysis)

No AA evaluation data available

LLM Stats Category Scores

(LLM Stats (zeroeval))
Long Context
80
Spatial Reasoning
80
Structured Output
80
Instruction Following
80
Frontend Development
80
Grounding
80
Multimodal
70
Search
70
Vision
70
Physics
60
Reasoning
60
General
60
Biology
60
Chemistry
60
Code
60
Math
50
Agents
50
Coding
50
Tool Calling
50

Pricing

Input Price$0.35 / 1M tokens
Output Price$1.5 / 1M tokens
Blended Price (3:1)$0.6375 / 1M tokens
Cache Read Price$0.04 / 1M tokens

Speed

No speed data available

Provider Price Ranking

Provider Price Ranking

5 providers

Cheapest: Kilo GatewayMost Expensive: Vercel AI Gateway
ProviderInputOutput
1Kilo GatewayCheapest
$0.3
$1.2
2MetaPRIMARY
$0.35
$1.5
3NanoGPT
$0.35
$1.5
4OpenRouter
$0.35
$1.5
5Vercel AI Gateway
$0.35
$1.5

Compare pricing across different API providers for this model.

External Sources