Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
STEM Knowledge on GPQA
Loading...
31.9
Pass@1 Accuracy
TAC
23.268
25.509
27.75
29.991
Jun 23, 2026
Pass@1 Accuracy
Updated 1mo ago
Evaluation Results
Method
Method
Links
Pass@1 Accuracy
TAC
Backbone=Qwen3-1.7B
2026.06
31.9
SEC
Backbone=Qwen3-1.7B
2026.06
30.7
Random
Backbone=Qwen3-1.7B
2026.06
30
TAC
Backbone=Llama3.2-3B
2026.06
28.7
Random
Backbone=Llama3.2-3B
2026.06
28
M2O
Backbone=Llama3.2-3B
2026.06
27.7
SEC
Backbone=Llama3.2-3B
2026.06
26.2
M2O
Backbone=Qwen3-1.7B
2026.06
25.8
Base
Backbone=Qwen3-1.7B
2026.06
24.2
Base
Backbone=Llama3.2-3B
2026.06
23.6
Feedback
Search any
task
Search any
task