The AI coding agent space just got a new entrant. Meta has launched an early beta of Muse Code, a terminal-based coding tool and the company’s first AI coding agent.
What Muse Code Is
Muse Code runs in the terminal and is powered by Muse Spark 1.2, Meta’s latest Muse Spark model. According to Meta’s launch blog post, the model was trained on long-horizon coding tasks, including whole-repository generation, large end-to-end projects, and auto-research.
How Meta Is Positioning It
Meta is pitching Muse Code as a more affordable alternative to what’s currently on the market, with Anthropic’s Claude Code and OpenAI’s Codex named as the direct comparisons. Exact pricing details were not disclosed in the announcement.
The Operator Takeaway
If you’re already running Claude Code or Codex and cost is a factor, Muse Code is worth watching. It’s early beta, so production use carries the usual caveats. The long-horizon training focus is the differentiator to track as benchmarks start appearing.
