Paweł Huryn: Muse Wonder Full fixed 44 of 105 real bugs, one behind GPT-6 Astra and ahead of Fable 5.1 and Opus 5.5
Paweł Huryn @PawelHurynX
Why it matters
An independent, real-repository test of Muse Wonder Full on the day after launch (about 42k views): it places Meta’s model level with OpenAI’s and Anthropic’s best, in contrast to early ‘benchmaxxed’ complaints.
Summary
Posted Oct 13, 2026 (16:34 UTC), opening a thread. Huryn says he tested Muse Wonder Full 1.0 “on real work. 2 repos. 105 bugs frontier models missed in Q1 2026. Find and fix what you can.” His results (bugs fixed, with an image of the table): GPT-6 Astra (max) 45; Muse Wonder Full 1.0 (max) 44; Fable 5.1 (max) 43; Opus 5.5 (max) 41.7; Muse Spark 1.3 (max) 32.2; Grok 4.7 (xhigh) 29. “@AIatMeta has just matched Anthropic and OpenAI.” He promises more effort levels and more runs later in the thread.
Caveats: a private test set by one person; the repositories, the grading and the number of runs per model are not described in this post (fractional scores suggest averages over runs); the rest of the thread and the image were not read for this file.
Metrics as of 2026-10-13 via api.fxtwitter.com: 41,826 views, 272 likes. Author: about 134,700 followers.
Archived text
So, I tested Muse Wonder Full 1.0 on real work. 2 repos. 105 bugs frontier models missed in Q1 2026. Find and fix what you can.
I’m so surprised we need a new thread. The results:
GPT-6 Astra (max): 45 Muse Wonder Full 1.0 (max): 44 Fable 5.1 (max): 43 Opus 5.5 (max): 41.7 Muse Spark 1.3 (max): 32.2 Grok 4.7 (xhigh): 29
@AIatMeta has just matched Anthropic and OpenAI. Huge congrats!
Next in this thread, more effort levels and more runs for Muse Wonder Full 1.0. All dropping tonight 🧵
