Meta says Muse Spark 1.3 has frontier performance — but its best results come from a model developers can’t broadly use yet

Meta's Muse Spark 1.3: Frontier Performance with a Catch

VentureBeat | The Independent Voice on AI

Carl Franzen | 9:19 am, PT, September 3, 2026

Credit: VentureBeat made with OpenAI ChatGPT-Images-2.0

Meta’s newest AI model, Muse Spark 1.3, unveiled yesterday, boasts frontier performance and is faster than its predecessor, Muse Spark 1.2. However, its best results come from a configuration (max reasoning) not yet broadly available to developers.

Key Takeaways:

  • Impressive Gains: Muse Spark 1.3 makes significant advancements, particularly in long-running agent tasks.
  • Two Configurations: Meta offers both max reasoning (stronger, but in testing) and xhigh reasoning (more widely available).
  • Competitive Benchmarking: While max reasoning leads in some tests, other models like Claude Opus and Claude Fable still outperform it overall.
  • Easier to Use: Meta claims Muse Spark 1.3 is easier to operate, with improved capabilities for context gathering, gap detection, and user clarification.

The Real Question:

The crucial question for companies considering deploying Muse Spark 1.3 is not just its potential for frontier performance but how accessible and affordable it truly is.