Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Agentic Task on ResearchQA
Loading...
73.7
Score
DR-Rubric-8B (GPT-5)
62.676
65.538
68.4
71.262
May 31, 2026
Score
Updated 1d ago
Evaluation Results
Method
Method
Links
Score
DR-Rubric-8B (GPT-5)
Training=SFT+RL, 1K
2026.05
73.7
DR-Rubric-8B (BS-3)
Training=SFT+RL, 3K
2026.05
72.4
DR-Rubric-8B (Gemini)
Training=SFT+RL, 1K
2026.05
71.7
Qwen3-8B-SFT
Training=SFT, 1K
2026.05
69.9
DR-Tulu-RL-8B
Training=SFT+RL, 25K
2026.05
67.1
DR-Tulu-SFT-8B
Training=SFT, 16K
2026.05
66.7
Qwen3-8B
2026.05
66.6
WebExplorer-8B
Training=SFT+RL, 25K
2026.05
66.1
Qwen2.5-7B
2026.05
65.5
Search-R1-7B
Training=RL, 90K
2026.05
63.1
Feedback
Search any
task
Search any
task