Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

LLM Agent Task Completion on τ-Bench (Retail & Airline Average)

18.1Success Rate (%)

DeepSeek-V3.1

11.65213.3261516.674Jun 2, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2026.06
18.1213.924.9
2026.06
1822.3022.3
2026.06
17.823.81.325.1
2026.06
17.4213.824.8
2026.06
16.920.53.924.4
2026.06
16.324.25.129.3
2026.06
15.424.2024.2
2026.06
11.923.12.725.8