Meta Launches Muse Code Beta: A Terminal-Based AI Coding Agent
The tool is powered by the new Muse Spark 1.2 model and is available for macOS and Linux. It aims to compete with offerings from Anthropic and OpenAI.
Meta has introduced Muse Code, a terminal-based AI coding agent now available in beta for macOS and Linux users. The tool is powered by Meta’s latest AI model, Muse Spark 1.2, and is designed to assist developers with complex software engineering tasks across large code repositories. This marks Meta’s entry into the AI coding-agent space, positioning it as a competitor to tools like Anthropic’s Claude Code and OpenAI’s Codex.
Muse Code is part of Meta’s broader AI initiatives, led by Meta Superintelligence Labs. The tool is available through a beta release, with installation instructions provided as a simple one-line command for users. The development team, including Meta’s AI chief Alexandr Wang, has emphasized the tool’s potential to streamline workflows and reduce costs for developers working on large-scale projects.
The Muse Spark 1.2 model, which powers Muse Code, is an upgrade from the previous Muse Spark 1.1 version. It is designed to improve the accuracy and efficiency of code generation and debugging. Developers can access Muse Code through a pay-as-you-go plan, with the standard tier priced at $1.25 per million input tokens and $4.25 per million output tokens, according to Meta’s pricing structure.
The release of Muse Code could shift the competitive landscape in AI coding tools, potentially influencing market dynamics and developer adoption. As a new entrant, Meta may face challenges related to vendor lock-in and integration with existing development ecosystems. The tool’s performance and cost structure could also impact its appeal to both individual developers and enterprise users.
Muse Code is still in its early development phase, with Meta indicating that further improvements and features are expected in the future. The tool’s beta release allows developers to test its capabilities and provide feedback, which could shape its evolution. As the AI coding-agent market continues to grow, Muse Code’s success will depend on its ability to meet user needs and differentiate itself from established competitors.
Sources
- https://9to5mac.com/2026/08/05/meta-launches-muse-code-ai-coding-agent-for-macos-and-linux/
- https://economictimes.indiatimes.com/tech/artificial-intelligence/meta-launches-new-ai-coding-tool-powered-by-muse-spark-1-2/articleshow/132979577.cms
- https://news.google.com/rss/articles/CBMioAFBVV95cUxQY2FRWTFiUk1MRHpON3ZwN3p2UTFqNXNOWWdjTDk5SjRMUk5RS3hEdXdyeUFLYkd5cU1idkhKVnZwM2FsYkhYalRFeTJwQUg4UjFtS0NrYzZLTHdQbGZLWGw4bTBwbU5UNnA1QWQwa2pVUWhkNWlzbTR2R3cydExpNE1EY2ZlQ2YxSzNrb0NqTGpuYldvaGxIVFVfY3FvVlk1?oc=5
- https://simonwillison.net/2026/Aug/5/muse-code-and-muse-spark-12/#atom-everything
- https://techcrunch.com/2026/08/05/meta-launches-muse-code-an-ai-agent-for-large-code-bases/
- https://www.engadget.com/2231285/meta-introduces-muse-code-its-take-on-a-coding-agent/