Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Mathematical Problem Solving on MATH 500 (Accuracy (avg@k))
Loading...
97.83
Accuracy (avg@k)
TRACE (FKL on Ky)
81.086
85.433
89.78
94.127
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
97.83
TRACE (RKL on Ey)
Backbone=Qwen3-8B, Sym...
2026.05
97.48
GRPO
Backbone=Qwen3-8B, Sym...
2026.05
97.3
Base
Backbone=Qwen3-8B, Sym...
2026.05
96.8
RLSD
Backbone=Qwen3-8B, Sym...
2026.05
96.68
SRPO
Backbone=Qwen3-8B, Sym...
2026.05
96.48
SDPO
Backbone=Qwen3-8B, Sym...
2026.05
95.13
TRACE (RKL on Ey)
Backbone=Qwen3-1.7B, S...
2026.05
92.05
Base
Backbone=Qwen3-1.7B, S...
2026.05
91.35
RLSD
Backbone=Qwen3-1.7B, S...
2026.05
91.12
TRACE (FKL on Ky)
Backbone=Qwen3-1.7B, S...
2026.05
91
GRPO
Backbone=Qwen3-1.7B, S...
2026.05
90.33
SRPO
Backbone=Qwen3-1.7B, S...
2026.05
86.02
SDPO
Backbone=Qwen3-1.7B, S...
2026.05
81.73
Feedback
Search any
task
Search any
task