Meta has launched Muse Code, a new AI-powered coding agent designed to help developers complete complex software engineering tasks across large codebases, as the company steps up competition in the rapidly expanding AI coding market.
Announcing the beta release, Meta CEO Mark Zuckerberg said Muse Code can handle complete software engineering workflows, including planning code changes, writing new code and validating results.
The terminal-based coding agent is powered by Muse Spark, Meta's previously released coding model, and can be installed with a single command.
Parallel AI agents for large projects
According to Zuckerberg, Muse Code manages large software repositories by automatically deploying multiple AI sub-agents that work simultaneously on different parts of a project.
"When a job is big enough, it fans out to separate sub-agents working in parallel in isolated worktrees," Zuckerberg said, adding that users' working copies remain untouched throughout the process.
He said internal testing showed the system successfully developed six separate game features simultaneously without conflicts between the AI agents.
Competing in the AI coding market
The launch positions Meta against rivals including OpenAI's Codex and Anthropic's Claude Code, which have gained popularity among developers seeking AI-assisted programming tools.
Meta's Chief AI Officer, Alexandr Wang, said the company aims to offer a cost-effective alternative for developers and businesses.
"We think that for a lot of workflows and a lot of use cases, this can be an incredibly good option, especially from a cost perspective," Wang told The Wall Street Journal.
Meta expands AI ambitions
The release of Muse Code is part of Meta's broader push to strengthen its AI portfolio.
In June, the company expanded beyond AI tools for advertising by launching enterprise AI agents focused on customer service and business support.
With Muse Code, Meta is seeking to establish a stronger presence in the fast-growing AI software development market, where companies are racing to build autonomous coding assistants capable of handling increasingly complex programming tasks.
