benchgap
Alibaba (Qwen) · model

Qwen3.8 27B (medium) benchmark scores

As of 2026-10-07, Qwen3.8 27B (medium) (Alibaba (Qwen)) has measured scores on 1 benchmark and estimated scores on 4 more.

BenchmarkScoreSource
τ³-Bench Banking47.4%measured
AA-AnalystAgent49.8%estimated ± 8.1 pp, low confidence
Harvey LAB94.2%estimated ± 2.1 pp, medium confidence
AutomationBench59.6%estimated ± 10.6 pp, low confidence
EnterpriseOps-Gym42.3%estimated ± 5.3 pp, medium confidence