Meta · Frontier
Muse Spark 1.3
Meta's Muse Spark 1.3 is a frontier multimodal reasoning model for long-horizon coding and agentic workflows, with strong gains in computer use, browsing, professional tool use, codebase understanding, instruction following, and million-token retrieval. It accepts text, images, audio, video, and files, 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
- Discovery tier
- Frontier
Pricing
USD per 1M tokens- Input
- $1.25
- Output
- $4.25
- Cache read
- $0.15
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 79% of 473 models
Side by side
Compare models
Put Muse Spark 1.3 next to up to three other models. The comparison stays in the link, so you can share it.
Public evaluations
Benchmarks
Independent evaluation results, with how each score ranks among the models in the Noumen catalog.
Intelligence Index
48.1
Tested at max reasoning effort
Better than 95% of 326 scored models
Indexes
Coding Index
75.8Composite of the coding evaluations
Better than 90% of 168 scored models
Agentic work
AutomationBench-AA
57.9%Workflow automation, penalized for guardrail violations
Better than 73% of 173 scored models
AutomationBench-AA tasks completed
29.8%Workflows finished end to end without a guardrail violation
Better than 84% of 173 scored models
Harvey LAB-AAv1.1
8.9%Legal tasks passed in full, with no material hallucination
Better than 96% of 29 scored models
AA-Briefcasev1.1
1,581 EloAgentic knowledge work, rated head to head
Better than 92% of 166 scored models
GDPval-AAv2.1
1,681 EloEconomically valuable occupational tasks, rated head to head
Better than 93% of 187 scored models
Document reasoning
GDP.pdf
26.6%Professional reasoning over PDF documents, all-pass rate
Better than 91% of 171 scored models
AA-LCRv1.1
83.0%Reasoning across long documents
Better than 90% of 309 scored models
MLCR-AA
13.9%Long-context reasoning over medical records
Better than 62% of 121 scored models
Reasoning and knowledge
Humanity’s Last Exam
48.7%Expert-level questions across academic fields
Better than 92% of 323 scored models
GPQA Diamond
93.5%Graduate-level science questions
Better than 92% of 291 scored models
Coding
Terminal-Benchv4.0
33.3%Practical software engineering in a terminal
Better than 80% of 171 scored models
SciCode
58.8%Scientific computing in Python
Better than 88% of 174 scored models
Legacy evaluations
AA-LCR (unversioned)
83.0%Long-context reasoning, earlier grading
Better than 90% of 309 scored models
Scores from Artificial Analysis, updated Oct 26, 2026. Muse Spark 1.3 was tested at max reasoning effort; compare scores measured at the same effort.Artificial Analysis
- #13Combined14.8%
Ranks among Noumen catalog models on Combined leaderboards. Only top-20 placements are listed.
- #18Roleplay18.9%
- #18Creative writing19.1%
- #9Coding25.8%
- #9Math19.4%
Ranks among Noumen catalog models on Use cases leaderboards. Only top-20 placements are listed.
- #9Overall1495.6
- #13Coding1535.8
- #9Math1509.8
- #16Longer Query1497.1
- #16Instruction Following1484.6
- #9Hard Prompts1517.2
- #16Chinese1531.2
- #7English1502.4
- #8Exclude Ties1513.7
- #13Expert1527.1
- #3French1525.2
- #1German1516.5
- #10Hard Prompts English1516.2
- #4Industry Business And Management And Financial Operations1500.6
- #17Industry Entertainment And Sports And Media1460.0
- #10Industry Legal And Government1502.3
- #12Industry Life And Physical And Social Science1512.0
- #18Industry Mathematical1499.8
- #15Industry Medicine And Healthcare1501.1
- #11Industry Software And It Services1528.0
- #16Industry Writing And Literature And Language1473.7
- #4Korean1492.8
- #13Multi Turn1493.7
- #10Non English1484.3
- #15Polish1485.2
- #11Russian1499.4
- #7Spanish1491.2
- #9Vision Style Control Overall1289.6
- #8Vision Style Control · Chinese1347.3
- #9Vision Style Control · Diagram1321.9
- #15Vision Style Control · English1279.3
- #16Vision Style Control · Humor1267.1
- #6Vision Style Control · Ocr1307.3
- #10WebDev Overall1657.2
- #8WebDev · Image To Webdev1640.4
- #10WebDev · Webdev1657.2
- #10WebDev · Webdev-html1652.7
- #10WebDev · Webdev-react1670.6
- #12Document Overall1470.7
- #15Document Style Control Overall1467.9
Ranks among Noumen catalog models on LMArena leaderboards. Only top-20 placements are listed.LMArena
- #3FrontierMath v2 (Tiers 1–3)74.0%
Ranks among Noumen catalog models on Epoch AI leaderboards. Only top-20 placements are listed.Epoch AI
- #16Creative Writing v31907
- #8Longform Writing82.8
Ranks among Noumen catalog models on EQ-Bench leaderboards. Only top-20 placements are listed.EQ-Bench
- #11Game Development1344
- #2Website1355
- #7UI Component1344
- #5Data Visualization1342
- #4SVG1320
- #73D1393
- #7ASCII Art1278
- #6Web Apps1297
- #4Full Stack1312
- #12Mobile Apps1220
Ranks among Noumen catalog models on Design Arena leaderboards. Only top-20 placements are listed.Design Arena
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
- 323.3 tok/s
- Faster than 94% of 257 models
- Time to first token
- 8.8 s
- Quicker than 13% of 257 models
- Prompt cache hits
- 23.6%
- About $0.99 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.
- 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.
curl https://www.noumen.xyz/api/v1/chat/completions \
-H "Authorization: Bearer $NOUMEN_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "meta/muse-spark-1.3",
"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-spark-1.3",
messages=[{"role": "user", "content": "Explain this stack trace."}],
)
print(reply.choices[0].message.content)Meta
Related models
Other Meta models, starting with the same family.