someaistuff.
metaReleased Oct 15, 2026

Meta: Muse Spark Open

OpenRouter lists a 1,048,576-token context and up to 943,718 output tokens for this open-weight model aimed at agentic coding.

Source description

Muse Spark Open is a coding-optimized open-weight model. Trained specifically for agentic coding workflows, it delivers higher first-attempt accuracy, reliable tool calling, and deep codebase understanding across the repos developers...

Compare
Context window1M
Max output943.7K
Provider endpoints8
Benchmark results19
Image inputAudio inputVideo inputTool callingStructured outputReasoningPrompt cachingOpen weights
Measured performance

Benchmark results

All leaderboards ↗
AA-Briefcase ↗1.1
1,279.25

Effort: xhigh · Xhigh

Independent result
GDPval-AA ↗2.1
1,401.12

Effort: xhigh · Xhigh

Independent result
AutomationBench-AA ↗AA Index 4.3.2 collection
49.71%

Effort: xhigh · Xhigh

Independent result
SciCode ↗AA Index 4.3.2 collection
58.45%AA · Under review

Effort: xhigh · Xhigh

Independent result
GDP.pdf ↗AA Index 4.3.2 collection
18.6%

Effort: xhigh · Xhigh

Independent result
CritPt ↗AA Index 4.3.2 collection
14.86%AA · Under review

Effort: xhigh · Xhigh

Independent result
AA-Omniscience ↗AA Index 4.3.2 collection
24.65

Effort: xhigh · Xhigh

Independent result
AA-LCR ↗1.1
78.33%

Effort: xhigh · Xhigh

Independent result
LiveBench Overall ↗2.0 · 2026-10-14 (dataset 2026-09-02)
57.57%

Effort: max · single run · equal mean of 7 suites

Independent result
LiveBench Design ↗2.0 · 2026-10-14 (dataset 2026-09-02)
60.3%

Effort: max · single run

Independent result
LiveBench Research ↗2.0 · 2026-10-14 (dataset 2026-09-02)
80.3%

Effort: max · single run

Independent result
LiveBench Browser Use ↗2.0 · 2026-10-14 (dataset 2026-09-02)
36%

Effort: max · single run

Independent result
LiveBench Coding ↗2.0 · 2026-10-14 (dataset 2026-09-02)
73.6%

Effort: max · single run

Independent result
LiveBench Agentic ↗2.0 · 2026-10-14 (dataset 2026-09-02)
18.9%

Effort: max · single run

Independent result
LiveBench Reasoning ↗2.0 · 2026-10-14 (dataset 2026-09-02)
73.7%

Effort: max · single run

Independent result
Access & cost

Prices across providers

8 quotes

Your token mix

USD · listed token rates

$2.3125
Together AIServerless · quantization=NVFP4
$2.3125
$2.3125
$2.3125
Nebius · via OpenRouternebius/fp4 · quantization=fp4
$2.3125
$2.3125
BaseTen · via OpenRouterbaseten/fp4 · quantization=fp4
$2.3125

OpenRouter

Catalog quote
Input
$1.25
Output
$4.25
Cached input
$0.15
USD / 1M tokens
1M context943.7K max outputPrice source ↗Checked Oct 17, 2026
Endpoint IDmeta/muse-spark-open

baseten

via Hugging Face
Input
$1.25
Output
$4.25
USD / 1M tokens
1M contextPrice source ↗Checked Oct 17, 2026
Endpoint IDmeta-models/Muse-Spark-Open-816B-A42B

Together AI

Serverless · quantization=NVFP4

Input
$1.25
Output
$4.25
Cached input
$0.15
USD / 1M tokens

Organization: Meta

1M contextPrice source ↗Checked Oct 17, 2026
Endpoint IDmeta-models/Muse-Spark-Open-816B-A42B

Modal

via OpenRouter

modal

Input
$1.25
Output
$4.25
Cached input
$0.15
USD / 1M tokens
1M context262.1K max outputPrice source ↗Checked Oct 17, 2026
Endpoint IDmeta/muse-spark-open::modal

Together

via OpenRouter

together

Input
$1.25
Output
$4.25
Cached input
$0.15
USD / 1M tokens
1M context943.7K max outputPrice source ↗Checked Oct 17, 2026
Endpoint IDmeta/muse-spark-open::together

Nebius

via OpenRouter

nebius/fp4 · quantization=fp4

Input
$1.25
Output
$4.25
USD / 1M tokens
1M context943.7K max outputPrice source ↗Checked Oct 17, 2026
Endpoint IDmeta/muse-spark-open::nebius/fp4

Fireworks

via OpenRouter

fireworks

Input
$1.25
Output
$4.25
Cached input
$0.15
USD / 1M tokens
1M context943.7K max outputPrice source ↗Checked Oct 17, 2026
Endpoint IDmeta/muse-spark-open::fireworks

BaseTen

via OpenRouter

baseten/fp4 · quantization=fp4

Input
$1.25
Output
$4.25
Cached input
$0.15
USD / 1M tokens
1M context943.7K max outputPrice source ↗Checked Oct 17, 2026
Endpoint IDmeta/muse-spark-open::baseten/fp4

Model specifications

Developer
meta
Modality
Language
Context
1,048,576 tokens
Maximum output
943,718 tokens
Weights
Open weights
Released
Oct 15, 2026
Provider listed
Sep 15, 2026
First discovered
Oct 16, 2026

Exact identities

  • openroutermeta/muse-spark-open
  • huggingfacemeta-models/Muse-Spark-Open-816B-A42B
  • artificial-analysis3661c8fb-caf7-4118-aa18-65b84fea8e44
  • livebenchmuse-spark-open

Sources

OpenRouter ↗Hugging Face ↗Artificial Analysis ↗LiveBench ↗

Updated Oct 17, 2026

More from meta

View all ↗
metaListed 5 days ago

Meta: Muse Wonder

Muse Wonder is a frontier reasoning model from Meta, designed for complex personal and professional agentic tasks. It accepts text, images, video, audio, and PDF documents and returns text, with...

1M contextImage inputAudio inputNo reported score
No reported score
$4.00in$20.00out/ 1M tokens · USDMeta · via OpenRouter · 2 endpointsmeta
+Compare
metaListed Sep 11

Meta: Muse Voice Transcribe 1.0

Muse Voice Transcribe 1.0 is a synchronous speech-to-text model from Meta. It is suited for push-to-talk, endpointing, and speaker-aware transcription, with keyword biasing for domain terms and language…

Audio inputNo reported score
No reported score
$50.00in$0.00out/ 1M tokens · USDOpenRouter
+Compare
metaListed Sep 2

Meta: Muse Spark 1.3 Contributor

Muse Spark 1.3 Contributor is the cost-efficient contributor tier of Meta’s multimodal reasoning model for experimentation, learning, and early-stage agentic, multi-agent, and coding workflows. It is…

1M contextImage inputVideo inputNo reported score
No reported score
$0.10in$0.20out/ 1M tokens · USDMeta · via OpenRouter · 2 endpointsmeta
+Compare
metaListed Sep 2

Meta: Muse Spark 1.3

Muse Spark 1.3 is a multimodal reasoning model from Meta for long-running agentic, multi-agent, and coding workflows. It is designed to keep track of information across extended tasks, work through...

1M contextImage inputVideo inputNo reported score
No reported score
$1.25in$4.25out/ 1M tokens · USDMeta · via OpenRouter · 2 endpointsmeta
+Compare