Muse Spark 1.3 tops DeepSWE at 75.4%, beating GPT-5.6 Sol and Fable 5

kimmonismus · x · 2026-09-03

Blogger kimmonismus notes that the newly released Muse Spark 1.3 takes first place on the DeepSWE coding benchmark with 75.4%, ahead of GPT-5.6 Sol and Fable 5. For reference, Fable 5 sits at 70%, and there appear to be no published evals yet for Fable 5.1.

Related event: Muse Spark 1.3 Tops DeepSWE Benchmark at 75.4%(2 posts)→

Original post →

More from Models

Models channel →