Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Knowledge-intensive Question Answering on GPQA (Accuracy)
Loading...
48.5
Accuracy
DIVER
19.068
26.709
34.35
41.991
May 12, 2026
Accuracy
Updated 21d ago
Evaluation Results
Method
Method
Links
Accuracy
DIVER
Backbone=Qwen3-4B
2026.05
48.5
GCPO
Backbone=Qwen3-4B
2026.05
47.5
DQO
Backbone=Qwen3-4B
2026.05
47.2
DAPO
Backbone=Qwen3-4B
2026.05
46
Div-R1
Backbone=Qwen3-4B
2026.05
45.2
GRPO
Backbone=Qwen3-4B
2026.05
44.4
GCPO
Backbone=Qwen3-1.7B
2026.05
31.8
DIVER
Backbone=Qwen3-1.7B
2026.05
30.2
DQO
Backbone=Qwen3-1.7B
2026.05
28.3
DAPO
Backbone=Qwen3-1.7B
2026.05
28.2
Qwen3-4B(Base)
Backbone=Qwen3-4B
2026.05
26.3
Div-R1
Backbone=Qwen3-1.7B
2026.05
25.1
GRPO
Backbone=Qwen3-1.7B
2026.05
24.3
Qwen3-1.7B(Base)
Backbone=Qwen3-1.7B
2026.05
20.2
Feedback
Search any
task
Search any
task