Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Agentic Tool-use on τ²-bench airline v1 (test)
Loading...
53.5
Pass@1
RL
42.06
45.03
48
50.97
Jun 29, 2026
Pass@1
Pass@4
Updated 25d ago
Evaluation Results
Method
Method
Links
Pass@1
Pass@4
RL
Backbone=Qwen3-8B, Pro...
2026.06
53.5
33.6
On-policy Distill.
Backbone=Qwen3-8B, Pro...
2026.06
52.6
32.9
Off-policy Distill.
Backbone=Qwen3-8B, Pro...
2026.06
51.8
31.3
Off. + On.
Backbone=Qwen3-8B, Pro...
2026.06
51.4
33.2
Off-policy Distill.
Backbone=Qwen3-8B, Pro...
2026.06
50
32.1
Pre-RL
Backbone=Qwen3-8B, Pro...
2026.06
42.5
26.5
Feedback
Search any
task
Search any
task