Summary
On August 5, 2026, Meta released Muse Code, a multi-agent coding harness with event-log replay, in beta, together with an updated Muse Spark 1.2 coding model. Muse Spark 1.2 offers a 1M-token context, async and parallel tool calls, and input pricing around $1.25 per million tokens. Muse Code is designed to handle end-to-end software engineering workflows including planning, writing code, and validating results.
What changed
Meta shipped Muse Code, a beta multi-agent coding harness with event-log replay, and the Muse Spark 1.2 coding model with 1M-token context, async/parallel tool calls, and ~$1.25/Mtok input pricing.
Why it matters
Meta re-entering agentic coding with both a harness and a competitively priced model puts a well-capitalized open-weight-friendly player against Anthropic, OpenAI, and coding-tool incumbents. Aggressive token pricing and a full harness aim to make Muse a default option for cost-sensitive coding-agent workloads.
Evidence excerpt
Meta launched Muse Code, its first AI coding agent, alongside the updated Muse Spark 1.2 coding model. Muse Code is designed to handle end-to-end software engineering workflows, including planning, writing code and validating results.