Meta's Muse Spark 1.3 Is Winning Creative Writing Evals by Refusing to Have a 'House Style'

In fresh subjective testing against GPT-6-Astra, Fable 5.1, and Gemini 3.8 flash, one evaluator strongly preferred Meta's Muse Spark for a counterintuitive reason: it does what you ask without imposing its own voice.

The frontier creative-writing race just produced a surprising leader. In new subjective testing across GPT-6-Astra, Fable 5.1, Muse Spark 1.3, and Gemini 3.8 flash, @sam_paech reported strongly preferring Meta's Muse Spark 1.3 over both GPT-6 and Fable — a notable result given how much attention the OpenAI and Fable lines have absorbed.

Unlock the full briefing

Get every story in today's briefing, the full archive, and the daily AI intelligence brief.

All stories today

Full archive

Daily brief

Cancel anytime. Payments powered by Stripe.