Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Science Question Answering on GPQA Diamond (avg@k)
Loading...
58.33
Accuracy (avg@k)
TRACE (FKL on Ky)
26.1732
34.5216
42.87
51.2184
May 11, 2026
Accuracy (avg@k)
Updated 2mo ago
Evaluation Results
Method
Method
Links
Accuracy (avg@k)
TRACE (FKL on Ky)
Backbone=Qwen3-8B, Sym...
2026.05
58.33
Base
Backbone=Qwen3-8B, Sym...
2026.05
58.27
RLSD
Backbone=Qwen3-8B, Sym...
2026.05
57.26
TRACE (RKL on Ey)
Backbone=Qwen3-8B, Sym...
2026.05
56.44
GRPO
Backbone=Qwen3-8B, Sym...
2026.05
53.85
SDPO
Backbone=Qwen3-8B, Sym...
2026.05
39.9
RLSD
Backbone=Qwen3-1.7B, S...
2026.05
38.89
TRACE (RKL on Ey)
Backbone=Qwen3-1.7B, S...
2026.05
38.67
SRPO
Backbone=Qwen3-8B, Sym...
2026.05
37.31
Base
Backbone=Qwen3-1.7B, S...
2026.05
37.18
GRPO
Backbone=Qwen3-1.7B, S...
2026.05
37.18
TRACE (FKL on Ky)
Backbone=Qwen3-1.7B, S...
2026.05
36.77
SRPO
Backbone=Qwen3-1.7B, S...
2026.05
35.82
SDPO
Backbone=Qwen3-1.7B, S...
2026.05
27.41
Feedback
Search any
task
Search any
task