News in Short

• Meta launches Muse Code AI coding agent in beta.
• The tool is designed for large software codebases.
• Muse Code can plan, write and validate code changes.
• It uses multiple AI sub-agents to work on tasks simultaneously.
• Meta positions it as a cost-effective alternative to Codex and Claude Code.

Meta has launched Muse Code, a new AI-powered coding agent designed to help developers work across large software projects. Available in beta, the tool can plan changes, write code and validate results while handling complex codebases using multiple AI agents working in parallel.

The launch marks Meta’s latest push into developer-focused AI tools as it takes on rivals such as OpenAI’s Codex and Anthropic’s Claude Code.

Built for complex software projects

Unlike AI coding assistants that mainly generate snippets of code, Muse Code is designed to complete end-to-end software engineering tasks.

According to Meta CEO Mark Zuckerberg, the AI can plan development work, write code, validate outputs and manage changes across large repositories.

The tool installs with a single command and runs directly from the terminal.

Multiple AI agents work in parallel

One of Muse Code’s standout features is its multi-agent architecture.

When assigned a large project, the AI automatically splits the workload into multiple sub-agents. Each operates inside an isolated worktree, allowing several features to be developed simultaneously without interfering with a developer’s existing code.

Meta says this approach improves efficiency while reducing conflicts during development.

Meta challenges AI coding rivals

Muse Code is powered by Meta’s previously released Muse Spark coding model.

With the launch, Meta joins a growing list of companies building AI software engineering tools, including OpenAI’s Codex and Anthropic’s Claude Code.

Meta’s AI chief Alexandr Wang said the company believes Muse Code offers a strong balance between performance and cost, making it an attractive option for many developer workflows.

Part of Meta’s broader AI strategy

The release comes as Meta continues expanding beyond consumer AI products.

In recent months, the company has increased investments in enterprise AI, including customer support agents and productivity tools, as it looks to strengthen its position in the rapidly growing AI software market.