List your company
ResearchFormerly WizWand

Mathematical Reasoning on AMC 24 (Pass@128)

79Pass@128Qwen2.5-7B + GRPO w/ VERL.
22.94437.49752.0566.603Sep 28, 2025
Updated 6mo ago

Evaluation results

MethodLinks
Qwen2.5-7B + GRPO w/ VERL.Base Model=Qwen2.5-7B, RL Framework=GRPO, VERL Integration=true, Decoding Setting=Pass@1282025.0979
Qwen2.5-7B + GRPO w/ VERL.k=128, Base Model=Qwen2.5-7B, RL Method=GRPO, Framework=VERL2025.0979
Qwen2.5-7B + GRPOBase Model=Qwen2.5-7B, RL Framework=GRPO, Decoding Setting=Pass@1282025.0978.3
Qwen2.5-7B + GRPOk=128, Base Model=Qwen2.5-7B, RL Method=GRPO2025.0978.3
Mathstral-7B-v0.1 + GRPO w/ VERL.k=128, Base Model=Mathstral-7B-v0.1, RL Method=GRPO, Framework=VERL2025.0976.9
Qwen2.5-7B + PPO w/ VERL.Base Model=Qwen2.5-7B, RL Framework=PPO, VERL Integration=true, Decoding Setting=Pass@1282025.0974.4
Qwen2.5-7B + PPO w/ VERL.k=128, Base Model=Qwen2.5-7B, RL Method=PPO, Framework=VERL2025.0974.4
Qwen2.5-7B + PPOBase Model=Qwen2.5-7B, RL Framework=PPO, Decoding Setting=Pass@1282025.0974.3
Qwen2.5-7B + PPOk=128, Base Model=Qwen2.5-7B, RL Method=PPO2025.0974.3
Qwen2.5-7BBase Model=Qwen2.5-7B, RL Framework=None, Decoding Setting=Pass@1282025.0973.7
Qwen2.5-7Bk=128, Base Model=Qwen2.5-7B2025.0973.7
Mathstral-7B-v0.1 + PPOk=128, Base Model=Mathstral-7B-v0.1, RL Method=PPO2025.0970.7
Mathstral-7B-v0.1 + PPO w/ VERL.k=128, Base Model=Mathstral-7B-v0.1, RL Method=PPO, Framework=VERL2025.0969.9
Mathstral-7B-v0.1 + GRPOk=128, Base Model=Mathstral-7B-v0.1, RL Method=GRPO2025.0969.2
Qwen2.5-3Bk=128, Base Model=Qwen2.5-3B2025.0969
Qwen2.5-3B + GRPOk=128, Base Model=Qwen2.5-3B, RL Method=GRPO2025.0968.5
Qwen2.5-3B + PPOk=128, Base Model=Qwen2.5-3B, RL Method=PPO2025.0967.9
Qwen2.5-3B + GRPO w/ VERL.k=128, Base Model=Qwen2.5-3B, RL Method=GRPO, Framework=VERL2025.0967.8
Qwen2.5-3B + PPO w/ VERL.k=128, Base Model=Qwen2.5-3B, RL Method=PPO, Framework=VERL2025.0967.3
Llama-3.1-8B-Instruct + GRPO w/ VERL.k=128, Base Model=Llama-3.1-8B-Instruct, RL Method=GRPO, Framework=VERL2025.0963.1
Mathstral-7B-v0.1k=128, Base Model=Mathstral-7B-v0.12025.0960.9
Llama-3.2-3B-Instruct + GRPOBase Model=Llama-3.2-3B-Instruct, RL Framework=GRPO, Decoding Setting=Pass@1282025.0960.6
Llama-3.2-3B-Instruct + GRPOk=128, Base Model=Llama-3.2-3B-Instruct, RL Method=GRPO2025.0960.6
Llama-3.1-8B-Instruct + PPO w/ VERL.k=128, Base Model=Llama-3.1-8B-Instruct, RL Method=PPO, Framework=VERL2025.0960
Llama-3.2-3B-Instruct + GRPO w/ VERL.Base Model=Llama-3.2-3B-Instruct, RL Framework=GRPO, VERL Integration=true, Decoding Setting=Pass@1282025.0959
Llama-3.2-3B-Instruct + GRPO w/ VERL.k=128, Base Model=Llama-3.2-3B-Instruct, RL Method=GRPO, Framework=VERL2025.0959
Llama-3.1-8B-Instruct + PPOk=128, Base Model=Llama-3.1-8B-Instruct, RL Method=PPO2025.0959
Llama-3.2-3B-Instruct + PPO w/ VERL.Base Model=Llama-3.2-3B-Instruct, RL Framework=PPO, VERL Integration=true, Decoding Setting=Pass@1282025.0957.8
Llama-3.2-3B-Instruct + PPO w/ VERL.k=128, Base Model=Llama-3.2-3B-Instruct, RL Method=PPO, Framework=VERL2025.0957.8
Llama-3.1-8B-Instructk=128, Base Model=Llama-3.1-8B-Instruct2025.0957.4
Llama-3.2-3B-Instruct + PPOBase Model=Llama-3.2-3B-Instruct, RL Framework=PPO, Decoding Setting=Pass@1282025.0957
Llama-3.2-3B-Instruct + PPOk=128, Base Model=Llama-3.2-3B-Instruct, RL Method=PPO2025.0957
Llama-3.1-8B-Instruct + GRPOk=128, Base Model=Llama-3.1-8B-Instruct, RL Method=GRPO2025.0956.9
Llama-3.2-3B-InstructBase Model=Llama-3.2-3B-Instruct, RL Framework=None, Decoding Setting=Pass@1282025.0951.5
Llama-3.2-3B-Instructk=128, Base Model=Llama-3.2-3B-Instruct2025.0951.5
Mistral-7B-v0.3k=128, Base Model=Mistral-7B-v0.32025.0939.6
Mistral-7B-v0.3 + GRPO w/ VERL.k=128, Base Model=Mistral-7B-v0.3, RL Method=GRPO, Framework=VERL2025.0938
Mistral-7B-v0.3 + GRPOk=128, Base Model=Mistral-7B-v0.3, RL Method=GRPO2025.0936
Mistral-7B-v0.3 + PPO w/ VERL.k=128, Base Model=Mistral-7B-v0.3, RL Method=PPO, Framework=VERL2025.0930.1
Mistral-7B-v0.3 + PPOk=128, Base Model=Mistral-7B-v0.3, RL Method=PPO2025.0925.1