Share your thoughts, 1 month free Claude Pro on usSee more

Math on OlpBench

49.1Accuracy

Qwen2.5-72B

Updated 22d ago

Evaluation Results

Method	Links
Qwen2.5-72B 2026.05		49.1
Qwen2.5-32B 2026.05		48.8
Qwen2.5-14B 2026.05		45.1
FORGE-7B-NP 2026.05		32.1
FORGE-7B-MATH 2026.05		31.8
FORGE-7B-NP-MATH 2026.05		31.2
Qwen2.5-7B 2026.05		30
Qwen2.5-3B 2026.05		29.7
InternLM3-8B 2026.05		25.1
Qwen2.5-7B-SFT 2026.05		21.4
LLama3.1-8b 2026.05		17.7