Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Scientific Reasoning on GPQA D (Score)
Loading...
85.4
Score
Confident Decoding
-3.416
19.642
42.7
65.758
Jun 20, 2026
Score
Updated 1mo ago
Evaluation Results
Method
Method
Links
Score
Confident Decoding
Model=Qwen3.5-122B-A10...
2026.06
85.4
Last Layer Decoding
Model=Qwen3.5-122B-A10...
2026.06
83.3
Confident Decoding
Model=Qwen3.5-35B-A3B,...
2026.06
82.8
Confident Decoding
Model=Gemma-4-31B, Dec...
2026.06
80.8
Confident Decoding
Model=gpt-oss-120B, De...
2026.06
80.8
Last Layer Decoding
Model=Qwen3.5-27B, Dec...
2026.06
80
Confident Decoding
Model=Qwen3.5-27B, Dec...
2026.06
80
Last Layer Decoding
Model=Gemma-4-31B, Dec...
2026.06
76.8
Last Layer Decoding
Model=Qwen3.5-35B-A3B,...
2026.06
76.3
Last Layer Decoding
Model=gpt-oss-120B, De...
2026.06
76.3
Last Layer Decoding
Model=Qwen3.5-9B-Base,...
2026.06
67.7
Confident Decoding
Model=Qwen3.5-9B-Base,...
2026.06
65.7
Last Layer Decoding
Model=Qwen3.5-9B, Deco...
2026.06
64.6
Confident Decoding
Model=Qwen3.5-9B, Deco...
2026.06
62.1
Confident Decoding
Model=gpt-oss-20b, Dec...
2026.06
60.1
Last Layer Decoding
Model=gpt-oss-20b, Dec...
2026.06
57.1
Last Layer Decoding
Model=Qwen3.5-0.8B, De...
2026.06
0
Confident Decoding
Model=Qwen3.5-0.8B, De...
2026.06
0
Feedback
Search any
task
Search any
task