Meta 发布 Muse Spark Open 编码模型，在 NVIDIA GB300 NVL72 上扩展智能体编码工作流
原标题：Scale Agentic Coding Workflows with Meta’s Muse Spark Open on NVIDIA GB300 NVL72
AI 摘要
Meta 发布面向编码优化的 816B 混合专家模型 Muse Spark Open，激活参数 42B，支持 100 万 token 上下文窗口，并在 NVIDIA GB300 NVL72 上训练以适配智能体编码工作流。该模型具备多步任务执行、工具调用、子智能体协调和计算机使用能力，同时原生支持文本、图像、视频和音频感知。NVIDIA 提供 SGLang、vLLM、TensorRT LLM 等推理栈及 NeMo AutoModel、Megatron Bridge 训练工具，并发布 NVFP4 权重和 NIM 部署方案。
正文节选
Meta launches Muse Spark Open, a coding-optimized 816B mixture-of-experts (MoE) model with 42B active parameters and a 1M token context window. Trained on NVIDIA GB300 NVL72 Blackwell Ultra specifically for agentic coding workflows, it delivers higher first attempt accuracy, reliable tool calling, and deep codebase understanding across repositories. The 1M token context window holds an entire project, including complex dependencies, legacy code, and thousands of files without breaking a task int