Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Mathematics on AIME25 (Avg@16)
Loading...
69.2
Avg@16
TRD
26.9968
37.9534
48.91
59.8666
May 8, 2026
May 13, 2026
May 18, 2026
May 23, 2026
May 28, 2026
Jun 2, 2026
Jun 7, 2026
Avg@16
Updated 1mo ago
Evaluation Results
Method
Method
Links
Avg@16
TRD
Backbone=Qwen3-8B (w/...
2026.06
69.2
+Reverse KL w/ Top-K
Backbone=Qwen3-8B (w/...
2026.06
68.5
+Forward KL w/ Clip
Backbone=Qwen3-8B (w/...
2026.06
68.3
+Reverse KL
Backbone=Qwen3-8B (w/...
2026.06
68.2
Base
Backbone=Qwen3-8B (w/...
2026.06
66.7
+Forward KL
Backbone=Qwen3-8B (w/...
2026.06
65.4
TRD
Backbone=Qwen3-4B-Inst...
2026.06
49.4
+Forward KL w/ Clip
Backbone=Qwen3-4B-Inst...
2026.06
49.2
+Reverse KL w/ Top-K
Backbone=Qwen3-4B-Inst...
2026.06
49
+Forward KL
Backbone=Qwen3-4B-Inst...
2026.06
48.8
+Reverse KL
Backbone=Qwen3-4B-Inst...
2026.06
48.8
Base
Backbone=Qwen3-4B-Inst...
2026.06
48.2
AIPO
Model=Qwen2.5-32B-Inst...
2026.05
39.69
GRPO
Model=Qwen2.5-32B-Inst...
2026.05
34.48
SFT
Model=Qwen2.5-32B-Inst...
2026.05
31.26
Original
Model=Qwen2.5-32B-Inst...
2026.05
28.62
Feedback
Search any
task
Search any
task