Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Scientific Question Answering on GPQA Diamond (pass@1)
Loading...
65.2
pass@1
Qwen3
3.736
19.693
35.65
51.607
May 13, 2026
pass@1
Updated 20d ago
Evaluation Results
Method
Method
Links
pass@1
Qwen3
Model Category=Origina...
2026.05
65.2
Qwen3-8192
Model Category=Origina...
2026.05
49
LUFFY
Model Category=Off-Pol...
2026.05
38.9
TGPO
Model Category=On-Poli...
2026.05
37.9
TGPO-annealing
Model Category=On-Poli...
2026.05
37.4
KDRL
Model Category=On-Poli...
2026.05
35.4
TGPOR
Model Category=On-Poli...
2026.05
33.3
GRPO++
Model Category=On-Poli...
2026.05
32.3
Oat-Zero
Model Category=On-Poli...
2026.05
23.7
SimpleRL-Zero
Model Category=On-Poli...
2026.05
23.2
SFT
Model Category=Off-Pol...
2026.05
20.2
PRIME-Zero
Model Category=On-Poli...
2026.05
18.2
Qwen2.5-Math-7B
Model Category=Origina...
2026.05
11.1
OP Distill
Model Category=On-Poli...
2026.05
6.1
Feedback
Search any
task
Search any
task