Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Scientific Reasoning on GPQA-D (Pass@1, #Tok)
Loading...
68.2
Pass@1 Accuracy
DyCon
37.208
45.254
53.3
61.346
Jun 5, 2026
Pass@1 Accuracy
Token Count (#Tok)
Updated 1mo ago
Evaluation Results
Method
Method
Links
Pass@1 Accuracy
Token Count (#Tok)
DyCon
backbone=Qwen3-4B-Thin...
2026.06
68.2
6,205
Baseline
backbone=QwQ-32B
2026.06
67.7
7,732
DyCon
backbone=QwQ-32B
2026.06
67.7
5,699
Baseline
backbone=Qwen3-4B-Thin...
2026.06
66.2
9,210
DyCon
backbone=Qwen3-14B
2026.06
65.2
6,180
Baseline
backbone=Qwen3-14B
2026.06
65.1
7,654
DyCon
backbone=DeepSeek-R1-D...
2026.06
47
5,304
Baseline
backbone=DeepSeek-R1-D...
2026.06
38.4
7,518
Feedback
Search any
task
Search any
task