Meta ships Muse Code, a terminal agent — and trains Muse Spark 1.2 to build whole repositories
Meta released Muse Code in beta on Wednesday: a terminal coding agent built for large codebases, with asynchronous background agents and a local event log that makes sessions replayable and restart-safe. It runs on Muse Spark 1.2, a coding-focused update trained heavily on long-horizon tasks including whole-repository generation; in one case study the model spent over 1,000 tool calls and up to 24 hours optimizing GPU kernels. The model is available now in Muse Code and through the Meta Model API.
Why it mattersMeta is the last of the big US labs to ship its own terminal agent, going head-to-head with Claude Code and its rivals. Its angle — day-long autonomous runs and repo-scale generation — shows the coding-agent race moving past autocomplete toward hands-off engineering shifts.
Read in the app — free, in 9 languages
Related stories
Rumour: the anonymous model that just topped a coding benchmark, for free, is said to be Zhipu's unreleased flagship2026-08-21DeepSeek's cheap workhorse can now see — and on agent tasks that need eyes it says it is close to Anthropic's best
2026-08-21Nvidia is paying $6 billion for the machine that builds a rival's models — and hiring 109 of the people who ran it
2026-08-21Given four hours and a GPU to improve the way AI is trained, the best agent scored 0.25 out of 1 — and most never tried
2026-08-21ChatGPT can now read your iMessages and send them — and the setting that lets it skip asking is the one OpenAI warns about
2026-08-21