NOUMENOpen app
NOUMEN

Meta · Open weights

Muse Glimmer 30B

Added Aug 11, 2026Open weights

Compare models

Choose up to 3 models to compare.1 of 3 chosen

Loading models…

Use in the app

Meta's Muse Glimmer 30B is a dense, open-weight multimodal model distilled from Muse Spark for long-horizon agents and coding workflows. It supports multi-step reasoning, reliable tool use, failure recovery, image understanding, and more than 100 languages.

  • Reasoning
  • Tools
  • Vision
  • Structured output
Input
Text, Images
Output
Text
Reasoning effort
minimal · low · medium · high · xhigh
Discovery tier
Open weights

Pricing

USD per 1M tokens
Input
$0.35
Output
$1.50
Cache read
$0.04

Reference prices as published by the model provider. API example

Specifications

Context window
131.1K
Max output
—
Average output (7 days)
682 tokens
Longer than 61% of 473 models

Side by side

Muse Glimmer 30B compared

Specifications, prices, speed and benchmark scores next to each other. Benchmarks below show every compared model too.

Compare models

Choose up to 3 models to compare.1 of 3 chosen

Loading models…

Muse Glimmer 30BMetaThis modelMuse Spark OpenMetaRemove
ProviderMetaMeta
Discovery tierOpen weightsFrontier
AddedAug 11, 2026Oct 15, 2026
Limits
Context window131.07K1.05MBest
Average output682 tokens1.3K tokens
Pricing, USD per 1M tokens
Input$0.35Best$1.25
Output$1.50Best$4.25
Cache read$0.04Best$0.15
Performance
Throughput69.2 tok/s—
Time to first token0.9 s—
Request success (24h)100.0%—
Benchmark indexes
Intelligence Index17.538.9Best
Coding Index49.0—
Capabilities
ReasoningYesYes
ToolsYesYes
VisionYesYes
Structured outputYesYes
Reasoning effortminimal, low, medium, high, xhighlow, medium, high, xhigh

Public evaluations

Benchmarks

Independent evaluation results, with how each score ranks among the models in the Noumen catalog.

  • Muse Glimmer 30B
  • Muse Spark Open

Intelligence Index

  • Muse Glimmer 30B17.5
  • Muse Spark Open38.9

Muse Glimmer 30B tested at high reasoning effort

Muse Glimmer 30B: Better than 46% of 326 scored models

3.8Catalog models by score55.1

Indexes

  • Coding Index

    Composite of the coding evaluations

    • Muse Glimmer 30B: 49.0
    • Muse Spark Open: —

Agentic work

  • AutomationBench-AA

    Workflow automation, penalized for guardrail violations

    • Muse Glimmer 30B: 6.8%
    • Muse Spark Open: 49.7%Top

  • AutomationBench-AA tasks completed

    Workflows finished end to end without a guardrail violation

    • Muse Glimmer 30B: 0.8%
    • Muse Spark Open: 20.9%Top

  • AA-Briefcasev1.1

    Agentic knowledge work, rated head to head

    • Muse Glimmer 30B: 477 Elo
    • Muse Spark Open: 1,279 EloTop

  • GDPval-AAv2.1

    Economically valuable occupational tasks, rated head to head

    • Muse Glimmer 30B: 783 Elo
    • Muse Spark Open: 1,401 EloTop

Document reasoning

  • GDP.pdf

    Professional reasoning over PDF documents, all-pass rate

    • Muse Glimmer 30B: 10.0%
    • Muse Spark Open: 18.6%Top

  • AA-LCRv1.1

    Reasoning across long documents

    • Muse Glimmer 30B: 83.3%Top
    • Muse Spark Open: 78.3%

  • MLCR-AA

    Long-context reasoning over medical records

    • Muse Glimmer 30B: —
    • Muse Spark Open: 7.8%

Reasoning and knowledge

  • Humanity’s Last Exam

    Expert-level questions across academic fields

    • Muse Glimmer 30B: 22.0%
    • Muse Spark Open: 43.9%Top

  • GPQA Diamond

    Graduate-level science questions

    • Muse Glimmer 30B: 83.5%
    • Muse Spark Open: 89.0%Top

Coding

  • Terminal-Benchv4.0

    Practical software engineering in a terminal

    • Muse Glimmer 30B: 0.5%
    • Muse Spark Open: 8.1%Top

  • SciCode

    Scientific computing in Python

    • Muse Glimmer 30B: 44.9%
    • Muse Spark Open: 58.4%Top

Legacy evaluations

  • AA-LCR (unversioned)

    Long-context reasoning, earlier grading

    • Muse Glimmer 30B: 83.3%Top
    • Muse Spark Open: 78.3%

Scores from Artificial Analysis, updated Oct 26, 2026. Muse Glimmer 30B was tested at high reasoning effort; compare scores measured at the same effort.Artificial Analysis

Live routing data

Performance and reliability

Recent averages across requests routed for this model. Your own speed depends on prompt length, reasoning effort and load.

Throughput
69.2 tok/s
Faster than 47% of 256 models
Time to first token
0.9 s
Quicker than 96% of 256 models
Prompt cache hits
49.8%
About $0.20 per 1M input tokens with caching
Request success (24h)
100.0%
Reliable

Cache hits are the share of input tokens recently served from a prompt cache across all requests. They are not a forecast of your own hit rate.

Hourly request success · last 72 hoursLowest hour 95.4%
  • Healthy (98% or more)
  • Degraded (90–98%)
  • Down (under 90%)
  • No requests

Developers

Use it through the API

Noumen’s API is OpenAI-compatible: any Chat Completions client works with a base URL and a key.

Read the quickstart
curl https://www.noumen.xyz/api/v1/chat/completions \
  -H "Authorization: Bearer $NOUMEN_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "meta/muse-glimmer-30b",
    "reasoning_effort": "medium",
    "messages": [{ "role": "user", "content": "Explain this stack trace." }]
  }'
import os
from openai import OpenAI

client = OpenAI(base_url="https://www.noumen.xyz/api/v1", api_key=os.environ["NOUMEN_API_KEY"])
reply = client.chat.completions.create(
    model="meta/muse-glimmer-30b",
    messages=[{"role": "user", "content": "Explain this stack trace."}],
)
print(reply.choices[0].message.content)

Meta

Other Meta models, starting with the same family.