Kimi K3 Reportedly Scores Twice as High as Claude 3.5 on Autonomous Legal Tasks

togethercompute · x · 2026-08-07

According to Together Compute, Kimi K3 scores nearly twice as high as Claude Fable 5 (likely a typo or alias for Claude 3.5) on Harvey LAB-AA's hard autonomous legal tasks.

Related event: Kimi K3 Nearly Doubles Claude's Score in Complex Legal Tasks(2 posts)→

Original post →

More from Models

Models channel →