List your company
ResearchFormerly WizWand

Mathematical Reasoning on CMath (pass@1)

91.7Pass@1Qwen2.5-7B + GRPO
6.9428.94550.9572.955Oct 11, 2023Mar 18, 2024Aug 25, 2024Feb 1, 2025Jul 10, 2025Dec 17, 2025May 26, 2026
Updated 5mo ago

Evaluation results

MethodLinks
Qwen2.5-7B + GRPOBase Model=Qwen2.5-7B, RL Framework=GRPO, VERL Integration=false2025.0991.7
Qwen2.5-7B + GRPO w/ VERL.Base Model=Qwen2.5-7B, RL Framework=GRPO, VERL Integration=true2025.0990.7
Qwen2.5-7B + PPO w/ VERL.Base Model=Qwen2.5-7B, RL Framework=PPO, VERL Integration=true2025.0990.3
VCRDTeacher Model=Qwen2.5-Math-7B-Instruct, Student Model=Qwen2.5-Math-1.5B2026.0490.2
Qwen2.5-7B + PPOBase Model=Qwen2.5-7B, RL Framework=PPO, VERL Integration=false2025.0989.8
MTTeacher Model=Qwen2.5-Math-7B-Instruct2026.0489.8
DistillLM-2Teacher Model=Qwen2.5-Math-7B-Instruct, Student Model=Qwen2.5-Math-1.5B2026.0489.7
DistilLLMTeacher Model=Qwen2.5-Math-7B-Instruct, Student Model=Qwen2.5-Math-1.5B2026.0489.3
GKDTeacher Model=Qwen2.5-Math-7B-Instruct, Student Model=Qwen2.5-Math-1.5B2026.0488.8
ABKDTeacher Model=Qwen2.5-Math-7B-Instruct, Student Model=Qwen2.5-Math-1.5B2026.0488.5
KDTeacher Model=Qwen2.5-Math-7B-Instruct, Student Model=Qwen2.5-Math-1.5B2026.0488.2
Qwen2.5-7BBase Model=Qwen2.5-7B, RL Framework=None, VERL Integration=false2025.0986.7
GPT-42023.1086
KwaiYiiMath-HPA#Params=13B, Human preference alignment=true2023.1085.83
KwaiYiiMath#Params=13B, Human preference alignment=false2023.1085.33
Ernie Bot2023.1084.33
ChatGPT2023.1073.83
ChatGLM2#Params=6B2023.1068.36
Qwen2.5-DenseLLM=Qwen2.5, Model=Dense, Activated Parameters=1.5 B2026.0563.99
QWen#Params=7B2023.1063.16
MSStudent Model=Qwen2.5-Math-1.5B2026.0462
Qwen2.5-Dense2MoELLM=Qwen2.5, Model=Ours, Activated Parameters=1.2 B2026.0555.5
BaiChuan1#Params=13B2023.1051.33
WizardMath#Params=13B2023.1050.83
Llama-3.2-3B-Instruct + GRPO w/ VERL.Base Model=Llama-3.2-3B-Instruct, RL Framework=GRPO, VERL Integration=true2025.0946.2
Llama-3.2-3B-Instruct + PPO w/ VERL.Base Model=Llama-3.2-3B-Instruct, RL Framework=PPO, VERL Integration=true2025.0930.7
Llama-3.2-3B-Instruct + GRPOBase Model=Llama-3.2-3B-Instruct, RL Framework=GRPO, VERL Integration=false2025.0928.3
Llama2-Dense2MoELLM=Llama2, Model=Ours, Activated Parameters=5.7 B2026.0525.33
Llama-3.2-3B-Instruct + PPOBase Model=Llama-3.2-3B-Instruct, RL Framework=PPO, VERL Integration=false2025.0921.2
Llama2-DenseLLM=Llama2, Model=Dense, Activated Parameters=7.0 B2026.0516.6
Llama-3.2-3B-InstructBase Model=Llama-3.2-3B-Instruct, RL Framework=None, VERL Integration=false2025.0910.2