Meta has entered the AI coding agent competition. The company has officially launched Muse Code in beta, placing the tool in direct competition with Anthropic's Claude Code and OpenAI's Codex. Terminal-based coding agents have become one of the fastest-moving segments of AI tooling, as developers shift from autocomplete-style helpers toward agents that can carry out multi-step engineering work with limited supervision.

Built on the Muse Spark 1.2 model, Muse Code is a terminal-based AI coding agent designed for complex software engineering tasks. It handles planning, code modifications, testing, and multi-agent coordination through persistent background sub-agents, all from the command line.

Pricing designed to undercut the competition

The standard tier is priced at $1.25 per million input tokens and $4.25 per million output tokens through the Meta Model API. A "contributor" tier costs roughly 10 to 20 times less than the standard option, with the tradeoff that users grant Meta permission to use their usage data for model improvements. The API is designed to be compatible with existing SDKs from OpenAI and Anthropic, which lowers the switching cost for teams already built on rival tooling. Installation is available via the command line for macOS and Linux users.

Benchmarks paint a competitive picture

On Terminal-Bench 2.1, Muse Code scored 82.9%, placing second behind Claude Opus 5 and Claude Code, which achieved 86.7%. On the DeepSWE 1.1 benchmark, it ranked third overall with a score of 59.3%. As is common with vendor-published figures, independent verification of these results may take time as external evaluators test the tool.

Muse Code supports multi-agent orchestration, spinning up persistent sub-agents that work in the background on different parts of a problem simultaneously. It uses isolated Git worktrees for parallel work, allowing multiple agents to modify code without interfering with each other's changes. For reliability, the tool features crash recovery through an append-only local event log. Meta highlighted an internal study in which Muse Code handled over 1,000 tool calls across a continuous 24-hour period.

The AI coding arms race heats up

Development is led by Alexandr Wang at Meta Superintelligence Labs. The Muse Spark model family first appeared in April 2026, with Muse Spark 1.1 arriving in July 2026. Muse Code is built on Muse Spark 1.2. The beta launch marks Meta's most direct push yet into a market where Anthropic and OpenAI have led in developer adoption, and further releases in the Muse Spark line will indicate how quickly Meta intends to iterate.