Meta launches Muse Code, an AI agent for large code bases

Meta, long considered a laggard in the AI tools space, is moving to catch up with rivals. This week, the company released a new terminal coding agent aimed at programmers working on large software code bases.

The tool, called Muse Code, is currently available in beta. Meta CEO Mark Zuckerberg said in a social media post that it can accomplish complete software engineering tasks across large repositories, including planning changes, writing code, and validating results. Muse Code can be installed with a single command and is powered by Meta's previously released coding model, Muse Spark.

To handle large projects, the system launches its own sub-agents that work simultaneously. Zuckerberg explained that when a job is big enough, it fans out to separate sub-agents working in parallel in isolated worktrees, so the user's working copy is never touched. In testing, Meta had it build six features for a game simultaneously with no collisions.

The move is intended to position Meta more competitively and affordably against AI lab peers such as OpenAI with its Codex coding agent and Anthropic with Claude Code. Alexandr Wang, Meta's AI chief who leads Meta Superintelligence Labs, told The Wall Street Journal that for many workflows and use cases, this could be an incredibly good option, especially from a cost perspective.

Meta has been expanding its AI presence through heavy investment. In June, it moved beyond its core focus of using AI to support advertising and entered the enterprise AI market with an agent aimed at customer service and support.