Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Agentic Terminal Tasks on Terminal-Bench out-of-distribution 2.0 (test)
Loading...
39.4
Avg@5
Tree-GRPO
25.568
29.159
32.75
36.341
May 11, 2026
Avg@5
Updated 22d ago
Evaluation Results
Method
Method
Links
Avg@5
Tree-GRPO
Base Model=Qwen3.5-35B...
2026.05
39.4
Tree-GRPO
Base Model=Nemotron-3-...
2026.05
37.2
Flat GRPO
Base Model=Qwen3.5-35B...
2026.05
34.2
Flat GRPO
Base Model=Nemotron-3-...
2026.05
33.8
Base
Base Model=Nemotron-3-...
2026.05
30.3
Base
Base Model=Qwen3.5-35B...
2026.05
26.1
Feedback
Search any
task
Search any
task