Meta AI Releases Muse Code (Beta): A Terminal Coding Agent Powered by the New Muse Spark 1.2 Model
Meta AI launched Muse Code, a beta terminal coding agent powered by the Muse Spark 1.2 model. The tool handles planning, writing, and validating code across large repositories using asynchronous background agents.
Meta Superintelligence Labs introduced Muse Code, a beta-stage agent designed for terminal-based development. It leverages the newly released Muse Spark 1.2 foundation model to execute complex coding tasks.
Unlike simple code completions, this agent manages entire workflows including planning changes, writing scripts, and validating results within large codebases. It utilizes asynchronous background agents that remain active throughout a user session.
This release signals Meta's continued investment in agentic AI workflows for software development. By focusing on terminal interaction and repository-wide validation, the tool aims to reduce manual overhead for developers working on extensive projects.
The Muse Spark 1.2 model underpins these capabilities, suggesting improvements in reasoning and context handling compared to previous iterations. As a beta product, it invites early feedback from the developer community before broader availability.
This page provides an editorial summary based on publicly available information. It is not a republished article. Use the source link below for the original report.