Meta · Frontier
Muse Spark Open
Built with Muse Spark Open. Meta's open-weight 816B-A42B model for coding and agentic workflows. Text and image input, text output, tool calling and structured output. Always reasons at low, medium, high, or xhigh effort. Custom Muse Model License and Usage Policy apply.
- Reasoning
- Tools
- Vision
- Structured output
- Input
- Text, Images
- Output
- Text
- Reasoning effort
- 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)
- 1.3K tokens
- Longer than 80% of 473 models
Side by side
Muse Spark Open compared
Specifications, prices, speed and benchmark scores next to each other. Benchmarks below show every compared model too.
| Muse Spark OpenMetaThis model | Grok 4.5xAIRemove | |
|---|---|---|
| Provider | Meta | xAI |
| Discovery tier | Frontier | Frontier |
| Added | Oct 15, 2026 | Jul 8, 2026 |
| Limits | ||
| Context window | 1.05MBest | 500K |
| Average output | 1.3K tokens | 1K tokens |
| Pricing, USD per 1M tokens | ||
| Input | $1.25Best | $2.00 |
| Output | $4.25Best | $6.00 |
| Cache read | $0.15Best | $0.30 |
| Performance | ||
| Throughput | — | 54.1 tok/s |
| Time to first token | — | 2.5 s |
| Request success (24h) | — | 100.0% |
| Benchmark indexes | ||
| Intelligence Index | 38.9Best | 38.8 |
| Coding Index | — | 72.4 |
| Capabilities | ||
| Reasoning | Yes | Yes |
| Tools | Yes | Yes |
| Vision | Yes | Yes |
| Structured output | Yes | Yes |
| Reasoning effort | low, medium, high, xhigh | low, medium, high |
Public evaluations
Benchmarks
Independent evaluation results, with how each score ranks among the models in the Noumen catalog.
- Muse Spark Open
- Grok 4.5
Intelligence Index
- Muse Spark Open38.9
- Grok 4.538.8
Muse Spark Open tested at xhigh reasoning effort
Muse Spark Open: Better than 83% of 326 scored models
Indexes
Coding Index
Composite of the coding evaluations
- Muse Spark Open: —
- Grok 4.5: 72.4
Agentic work
AutomationBench-AA
Workflow automation, penalized for guardrail violations
- Muse Spark Open: 49.7%
- Grok 4.5: 57.9%Top
AutomationBench-AA tasks completed
Workflows finished end to end without a guardrail violation
- Muse Spark Open: 20.9%
- Grok 4.5: 26.3%Top
AA-Briefcasev1.1
Agentic knowledge work, rated head to head
- Muse Spark Open: 1,279 EloTop
- Grok 4.5: 1,278 Elo
GDPval-AAv2.1
Economically valuable occupational tasks, rated head to head
- Muse Spark Open: 1,401 EloTop
- Grok 4.5: 1,381 Elo
Document reasoning
GDP.pdf
Professional reasoning over PDF documents, all-pass rate
- Muse Spark Open: 18.6%
- Grok 4.5: 18.8%Top
AA-LCRv1.1
Reasoning across long documents
- Muse Spark Open: 78.3%
- Grok 4.5: 79.3%Top
MLCR-AA
Long-context reasoning over medical records
- Muse Spark Open: 7.8%
- Grok 4.5: 16.1%Top
Reasoning and knowledge
Humanity’s Last Exam
Expert-level questions across academic fields
- Muse Spark Open: 43.9%Top
- Grok 4.5: 42.7%
GPQA Diamond
Graduate-level science questions
- Muse Spark Open: 89.0%
- Grok 4.5: 93.1%Top
Coding
Terminal-Benchv4.0
Practical software engineering in a terminal
- Muse Spark Open: 8.1%
- Grok 4.5: 10.6%Top
SciCode
Scientific computing in Python
- Muse Spark Open: 58.4%Top
- Grok 4.5: 55.0%
Legacy evaluations
AA-LCR (unversioned)
Long-context reasoning, earlier grading
- Muse Spark Open: 78.3%
- Grok 4.5: 79.3%Top
Scores from Artificial Analysis, updated Oct 26, 2026. Muse Spark Open was tested at xhigh reasoning effort; compare scores measured at the same effort.Artificial Analysis
- Muse Spark Open
- Grok 4.5
| Leaderboard | Muse Spark Open | Grok 4.5 |
|---|---|---|
| Website | #15 · 1306 | Not in top 20 |
Ranks among Noumen catalog models on Design Arena leaderboards. Only top-20 placements are listed.Design Arena
- Muse Spark Open
- Grok 4.5
| Leaderboard | Muse Spark Open | Grok 4.5 |
|---|---|---|
| Vision Style Control · Chinese | Not in top 20 | #16 · 1330.1 |
| Vision Style Control · Creative Writing Vision | Not in top 20 | #14 · 1280.9 |
| Search Overall | Not in top 20 | #10 · 1215.9 |
| Document Style Control Overall | Not in top 20 | #16 · 1464.3 |
Ranks among Noumen catalog models on LMArena leaderboards. Only top-20 placements are listed.LMArena
- Muse Spark Open
- Grok 4.5
| Leaderboard | Muse Spark Open | Grok 4.5 |
|---|---|---|
| FrontierCode Main | Not in top 20 | #20 · 42.4% |
| FrontierCode Extended | Not in top 20 | #18 · 56.5% |
Ranks among Noumen catalog models on FrontierCode leaderboards. Only top-20 placements are listed.FrontierCode
- Muse Spark Open
- Grok 4.5
| Leaderboard | Muse Spark Open | Grok 4.5 |
|---|---|---|
| ARC-AGI-3 (Default harness) | Not in top 20 | #16 · 0.3% |
Ranks among Noumen catalog models on ARC Prize leaderboards. Only top-20 placements are listed.ARC Prize
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.
- Prompt cache hits
- 0.0%
- About $1.25 per 1M input tokens with caching
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.
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-open",
"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-open",
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.