Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Mathematical Reasoning on MATH 500 (EM, #Tool)
Loading...
84.6
EM (Accuracy)
ParetoPO
65.256
70.278
75.3
80.322
Jun 15, 2026
EM (Accuracy)
Avg. Tool Usage
Updated 1mo ago
Evaluation Results
Method
Method
Links
EM (Accuracy)
Avg. Tool Usage
ParetoPO
Base Model=Qwen2.5-Mat...
2026.06
84.6
1.2
ToRL-GRPO
Base Model=Qwen2.5-Mat...
2026.06
82.2
2.2
OTC-GRPO
Base Model=Qwen2.5-Mat...
2026.06
81
1.5
MO-GRPO
Base Model=Qwen2.5-Mat...
2026.06
80.8
1.9
ParetoPO
Base Model=Qwen2.5-Mat...
2026.06
80
0.9
Qwen2.5-Math-7B-Ins-TIR
Base Model=Qwen2.5-Mat...
2026.06
78.8
1.6
ToRL-GRPO
Base Model=Qwen2.5-Mat...
2026.06
77.8
2.1
Qwen2.5-Math-7B-Ins
Base Model=Qwen2.5-Mat...
2026.06
74.8
-
OTC-GRPO
Base Model=Qwen2.5-Mat...
2026.06
74
1.3
Qwen2.5-Math-1.5B-Ins-TIR
Base Model=Qwen2.5-Mat...
2026.06
73.8
1.3
MO-GRPO
Base Model=Qwen2.5-Mat...
2026.06
71.2
2
Qwen2.5-Math-1.5B-Ins
Base Model=Qwen2.5-Mat...
2026.06
66
-
Feedback
Search any
task
Search any
task