Muse Spark 1.1 Leads Multiple Benchmarks

alexandr_wang · x · 2026-07-09

This repost claims that Muse Spark 1.1 achieves twice the performance of Fable on Harvey’s LegalBench, while simultaneously securing SOTA on Harvey’s LegalBench, TaxEval, and MedScribe.

Related event: Meta Launches Muse Spark 1.1: A Low-Cost, High-Performance Agentic Model(133 posts)→

Original post →

More from Models

Models channel →