Muse Spark 1.2 Contributor (Meta Contributor)

Muse Spark 1.2 Contributor sits as the budget-oriented contributor tier within the Muse Spark 1.2 family, aimed at developers who want to experiment, learn, or build early-stage projects without higher spend. According to its listing, it is framed as a reasoning model from Meta tailored for complex agentic tasks, designed to coordinate multi-agent workflows either as a main agent that plans and delegates or as a parallel subagent. The positioning emphasizes accessibility for new builders while still supporting sophisticated developer use cases such as whole-repository generation, multi-file refactors, and extended debugging sessions that stretch beyond a single prompt.

For practical fit, the listing describes broad input handling across text, images, video, audio, and PDF documents with text output, paired with a 1M-token context window that supports long-running agentic sessions. It is built to plug into multiple coding harnesses and offers structured output, parallel function calling, and configurable reasoning effort, giving developers fine-grained control over how the model plans and executes tasks. The note that prompts and outputs may be used to improve Meta's products reinforces its intended role as an experimentation and learning tier rather than a production-hardened variant.