Qwen3.8-27B Tops Harvey's Legal Agent Benchmark

Alibaba_Qwen · x · 2026-08-19

Qwen3.8-27B ranks #1 on Harvey's Legal Agent benchmark, tying with Fable 5 at 11.3 points and outperforming Kimi K3, Qwen 3.8 Max, and DeepSeek V4. The model highlights its capability for professional tasks while remaining small enough to run locally for everyday workflows.

Original post →

More from coding & agent

coding & agent channel →