MODEL PROFILE

Muse Spark Open (Xhigh)

Meta · muse-spark-open

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...

Text generationreasoningstructured_outputstools
Context window1,048,576 Token
Input → OutputText、Image、Audio、Video → Text
Model weightsOpen weights
Profile updated2026/11/7 00:57:38
OVERVIEW

Model overview

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...

The summary below is compiled from sources by an AI model, not our own testing. Verify selection decisions against original sources and your workload.

Key features

  • 专为代理式编程工作流(agentic coding workflows)优化,提升首次生成代码的准确率
  • 支持多模态输入(文本、图像、音频、视频),但仅输出文本
  • 具备可靠工具调用能力与深度代码库理解能力

Limitations & considerations

Sources are insufficient to establish specific limitations. Verify task performance, latency, data policies, licensing and version lifecycle.

PRICING & COST

Source prices & cost analysis

Prices vary by channel, region and configuration. Channel quotes are listed independently and are not treated as a single official provider price.

Source / channelInputOutputBilling unitCollected at
OpenRouter · meta/muse-spark-open ↗USD 1.25USD 4.25Per million tokens2026/11/2 06:25:21
Artificial Analysis · Xhigh ↗USD 1.25USD 4.25Per million tokens2026/11/2 06:30:46

Some channels offer cache discounts. Check the source for cache read, write and additional fees.

Estimate monthly base token costs

—

Estimates exclude cache discounts, tiered pricing, multimodal surcharges, web search, taxes and additional reasoning output.

INDEPENDENT BENCHMARKS

Public benchmarks & run configuration

Terminal-Bench 4.00.08XhighArtificial Analysis ↗采集于 2026/11/2 06:30:46
AA Intelligence Index24.68XhighArtificial Analysis ↗采集于 2026/11/2 06:30:46
Time to first token0.73XhighArtificial Analysis ↗采集于 2026/11/2 06:30:46
Output speed139.31XhighArtificial Analysis ↗采集于 2026/11/2 06:30:46
View full indexed rankings →

Frequently asked questions

Muse Spark Open (Xhigh) 是否支持多模态输入?

是,支持文本、图像、音频和视频四种输入模态。

该模型能否直接执行代码或调用外部工具?

资料明确标注其具备‘reliable tool calling’(可靠工具调用)能力,表明设计上支持工具调用,但不等于自主执行。

其输出是否支持结构化格式(如 JSON)?

是,标签中包含 'structured_outputs',说明支持结构化输出。

Do open weights permit free commercial use?

Not necessarily. Check the version license, commercial conditions and usage policies. Self-hosting incurs compute and maintenance costs.

Are these real-time billing prices?

These are timestamped source snapshots, not live billing prices. Check the channel for current versions, regions, discounts and billing methods.

Sources & traceability

许可证:Muse Model License。模型介绍与第三方评测均以原来源的具体版本和方法为准。