Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Tool-use on BFCL, MCP-Atlas, τ²-Bench, and VitaBench Summary
Loading...
33.4
Average Success Rate
EnvFactory
15.5848
20.2099
24.835
29.4601
May 18, 2026
Average Success Rate
Updated 2mo ago
Evaluation Results
Method
Method
Links
Average Success Rate
EnvFactory
Backbone=Qwen3-8B, Env...
2026.05
33.4
EnvScaler
Backbone=Qwen3-8B, Env...
2026.05
32.72
EnvFactory (SFT)
Backbone=Qwen3-8B, Env...
2026.05
30.82
EnvFactory
Backbone=Qwen3-4B, Env...
2026.05
30.77
EnvScaler
Backbone=Qwen3-4B, Env...
2026.05
29.56
Base
Backbone=Qwen3-8B, Env...
2026.05
29.23
AWM
Backbone=Qwen3-8B, Env...
2026.05
28.65
EnvFactory (SFT)
Backbone=Qwen3-4B, Env...
2026.05
27.29
AWM
Backbone=Qwen3-4B, Env...
2026.05
25.47
Base
Backbone=Qwen3-4B, Env...
2026.05
24.09
EnvFactory
Backbone=Qwen3-1.7B, E...
2026.05
19.74
EnvFactory (SFT)
Backbone=Qwen3-1.7B, E...
2026.05
18.6
EnvScaler
Backbone=Qwen3-1.7B, E...
2026.05
16.51
Base
Backbone=Qwen3-1.7B, E...
2026.05
16.27
Feedback
Search any
task
Search any
task