NOUMENOpen app
NOUMEN

Meta · Frontier

Muse Wonder

Added Oct 12, 2026

Compare models

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

Loading models…

Use in the app

Meta's Muse Wonder is a frontier reasoning model designed for complex personal and professional agentic tasks. It accepts text, images, audio, video, and PDF documents across a one-million-token context window, supports tool calling and structured output, and always reasons before answering.

  • Reasoning
  • Tools
  • Vision
  • Structured output
  • PDF input
  • Audio input
  • Video input
Input
Text, Images, Video, Audio
Output
Text
Reasoning effort
minimal · low · medium · high · xhigh · max
Discovery tier
Frontier

Pricing

USD per 1M tokens
Input
$4.00
Output
$20.00
Cache read
$0.20

Reference prices as published by the model provider. API example

Specifications

Context window
1M
Max output
—
Average output (7 days)
1.2K tokens
Longer than 77% of 473 models

Side by side

Muse Wonder 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 WonderMetaThis modelMuse Spark OpenMetaRemove
ProviderMetaMeta
Discovery tierFrontierFrontier
AddedOct 12, 2026Oct 15, 2026
Limits
Context window1.05M1.05M
Average output1.2K tokens1.3K tokens
Pricing, USD per 1M tokens
Input$4.00$1.25Best
Output$20.00$4.25Best
Cache read$0.20$0.15Best
Performance
Throughput102.6 tok/s—
Time to first token1.8 s—
Request success (24h)99.5%—
Benchmark indexes
Intelligence Index52.4Best38.9
Capabilities
ReasoningYesYes
ToolsYesYes
VisionYesYes
Structured outputYesYes
PDF inputYesNo
Audio inputYesNo
Video inputYesNo
Reasoning effortminimal, low, medium, high, xhigh, maxlow, medium, high, xhigh

Public evaluations

Benchmarks

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

  • Muse Wonder
  • Muse Spark Open

Intelligence Index

  • Muse Wonder52.4
  • Muse Spark Open38.9

Muse Wonder tested at max reasoning effort

Muse Wonder: Better than 97% of 326 scored models

3.8Catalog models by score55.1

Agentic work

  • AutomationBench-AA

    Workflow automation, penalized for guardrail violations

    • Muse Wonder: 68.1%Top
    • Muse Spark Open: 49.7%

  • AutomationBench-AA tasks completed

    Workflows finished end to end without a guardrail violation

    • Muse Wonder: 39.1%Top
    • Muse Spark Open: 20.9%

  • Harvey LAB-AAv1.1

    Legal tasks passed in full, with no material hallucination

    • Muse Wonder: 10.0%
    • Muse Spark Open: —

  • AA-Briefcasev1.1

    Agentic knowledge work, rated head to head

    • Muse Wonder: 1,659 EloTop
    • Muse Spark Open: 1,279 Elo

  • GDPval-AAv2.1

    Economically valuable occupational tasks, rated head to head

    • Muse Wonder: 1,770 EloTop
    • Muse Spark Open: 1,401 Elo

Document reasoning

  • GDP.pdf

    Professional reasoning over PDF documents, all-pass rate

    • Muse Wonder: 30.6%Top
    • Muse Spark Open: 18.6%

  • AA-LCRv1.1

    Reasoning across long documents

    • Muse Wonder: 80.7%Top
    • Muse Spark Open: 78.3%

  • MLCR-AA

    Long-context reasoning over medical records

    • Muse Wonder: 22.2%Top
    • Muse Spark Open: 7.8%

Reasoning and knowledge

  • Humanity’s Last Exam

    Expert-level questions across academic fields

    • Muse Wonder: 50.1%Top
    • Muse Spark Open: 43.9%

  • GPQA Diamond

    Graduate-level science questions

    • Muse Wonder: —
    • Muse Spark Open: 89.0%

Coding

  • Terminal-Benchv4.0

    Practical software engineering in a terminal

    • Muse Wonder: 52.0%Top
    • Muse Spark Open: 8.1%

  • SciCode

    Scientific computing in Python

    • Muse Wonder: 60.3%Top
    • Muse Spark Open: 58.4%

Legacy evaluations

  • AA-LCR (unversioned)

    Long-context reasoning, earlier grading

    • Muse Wonder: 80.7%Top
    • Muse Spark Open: 78.3%

Scores from Artificial Analysis, updated Oct 26, 2026. Muse Wonder was tested at max 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
102.6 tok/s
Faster than 63% of 257 models
Time to first token
1.8 s
Quicker than 78% of 257 models
Prompt cache hits
48.2%
About $2.17 per 1M input tokens with caching
Request success (24h)
99.5%
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 66.6%
  • 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-wonder",
    "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-wonder",
    messages=[{"role": "user", "content": "Explain this stack trace."}],
)
print(reply.choices[0].message.content)

Meta

Other Meta models, starting with the same family.