Meta Enters AI-Powered Coding Tools Market with Muse Code

Artificial intelligence (AI) is rapidly advancing in coding, and Meta is the latest tech giant to enter the space. On Wednesday, August 5, the Mark Zuckerberg-led company introduced Muse Code, a coding tool powered by its latest AI model, Muse Spark 1.2. The tool is designed to help developers write and debug software. With this launch, Meta is entering the AI-powered coding tools market, where companies such as OpenAI and Anthropic have already established a strong presence. “We’re releasing Muse Spark 1.2 and an early beta of Muse Code, a purpose-built coding agent optimised for long-horizon, multi-agentic coding workflows and transparent auditability: every subagent it spawns, every tool call, every steer and cancel, is observable and replayable through the event log,” the company said in a blog post. Meta said that the range of Muse Code’s capabilities include writing code and verifying results. The model has been trained to handle long, complex coding projects while simultaneously running multiple sub-agents to accelerate difficult tasks. Along with the new AI model, Meta also released Muse Code to developers. The company also said that it was expanding global access to its public review. This is to enable developers worldwide to start building with Muse Spark in Muse Code and Meta Model API. When it comes to performance, Meta says Muse Spark 1.2 delivers stronger coding performance than its predecessor across several software engineering benchmarks. According to the company, the model scored 82.9 per cent on Terminal-Bench 2.1, placing it close to Anthropic’s Claude Opus 5 (86.7 per cent), while also posting gains on Meta’s internal coding evaluations. The company said the model is optimised for long-running, agentic coding tasks, including debugging, feature implementation and code migration across large repositories. Meanwhile, Muse Code combines Muse Spark 1.2 with persistent background agents, parallel sub-agents and a crash-safe event log, allowing it to maintain context and resume complex software engineering workflows after interruptions. Muse Code arrives at a time when AI-powered software development is becoming one of the most competitive areas in generative AI. Worldwide, developers are rapidly moving beyond simple code completion to deploying AI agents that can understand entire codebases, write new features, debug errors, run tests, and accomplish complex engineering tasks, and all of these with minimal human supervision. At present, Anthropic’s Claude Code, OpenAI’s Codex, Google’s Gemini CLI and Microsoft’s GitHub Copilot are among the leading tools driving this shift, while AI-native coding platforms such as Cursor and Windsurf have also gained significant traction. The adopting of these tools at breakneck speeds has also pushed manual programming that involved writing every line of code to revewing, refining and organising AI-generated code. Instead of fomenting fears of job displacement among developers, the new line of AI coding agents are working as active collaborators on software engineering and are seen to be automating routine work thereby accelerating development cycles. Meta’s entry signals its ambition to compete in a fast-growing market where coding assistants are emerging as a key battleground for AI companies alongside chatbots and search.