Code Generation on LiveCodeBench (Avg)
168Average ScoreExpertWeaver
Evaluation results
| Method | Links | |
|---|---|---|
| ExpertWeaverBase Model=DeepSeek-R1-Distill-Qwen-7B, Sparsity=12.5%2026.02 | 168 | |
| DeepSeek-R1Model Category=Closed-source Models2026.01 | 83.8 | |
| o1-miniModel Category=Closed-source Models2026.01 | 75 | |
| Claude3.7-SonnetModel Category=Closed-source Models2026.01 | 63.2 | |
| LLM-PrunerBase Model=DeepSeek-R1-Distill-Qwen-7B, Sparsity=12.5%2026.02 | 61 | |
| GenRM-R-Align-8BReward Model=GenRM-R-Align-8B2026.02 | 51.5 | |
| LoGos(32B)Model Category=Open-source Models (32B), Parameter Count=32B2026.01 | 50.9 | |
| Qwen3-8BReward Model=Baseline2026.02 | 50.8 | |
| Qwen3-14B-as-GenRMReward Model=Qwen3-14B-as-GenRM2026.02 | 50.5 | |
| GenRM-RLVR-14BReward Model=GenRM-RLVR-14B2026.02 | 50.1 | |
| GenRM-R-Align-14BReward Model=GenRM-R-Align-14B2026.02 | 49.4 | |
| Qwen3-8B-as-GenRMReward Model=Qwen3-8B-as-GenRM2026.02 | 48.8 | |
| GenRM-RLVR-8BReward Model=GenRM-RLVR-8B2026.02 | 47.4 | |
| Qwen2.5-Coder-7B-InstructSize=7B2024.11 | 37.6 | |
| DeepSeek-R1-Distill-Qwen-32BModel Category=Open-source Models (32B), Parameter Count=32B, Model Variant=Distill2026.01 | 36.5 | |
| IOAModel=Qwen2.5-14B, Teacher Model=OpenAI o12026.02 | 30.63 | |
| Qwen2.5-32B-InstructModel Category=Open-source Models (32B), Parameter Count=32B, Model Variant=Instruct2026.01 | 29.9 | |
| IOAModel=Qwen2.5-7B, Teacher Model=OpenAI o12026.02 | 29.54 | |
| MADAModel=Qwen2.5-14B, Teacher Model=OpenAI o12026.02 | 26.76 | |
| Star-AgentsModel=Qwen2.5-14B, Teacher Model=OpenAI o12026.02 | 26.23 | |
| IOAModel=LLaMA3.1-8B, Teacher Model=OpenAI o12026.02 | 26.08 | |
| MADAModel=Qwen2.5-7B, Teacher Model=OpenAI o12026.02 | 25.66 | |
| CasCoDModel=Qwen2.5-14B, Teacher Model=OpenAI o12026.02 | 25.34 | |
| Star-AgentsModel=Qwen2.5-7B, Teacher Model=OpenAI o12026.02 | 25.13 | |
| CounterDistillModel=Qwen2.5-14B, Teacher Model=OpenAI o12026.02 | 24.99 | |
| DSSModel=Qwen2.5-14B, Teacher Model=OpenAI o12026.02 | 24.55 | |
| DS-Coder-V2-Lite-InstructSize=16B2024.11 | 24.3 | |
| CasCoDModel=Qwen2.5-7B, Teacher Model=OpenAI o12026.02 | 24.24 | |
| CounterDistillModel=Qwen2.5-7B, Teacher Model=OpenAI o12026.02 | 23.89 | |
| DSSModel=Qwen2.5-7B, Teacher Model=OpenAI o12026.02 | 23.45 | |
| LionModel=Qwen2.5-14B, Teacher Model=OpenAI o12026.02 | 23.44 | |
| Yi-Coder-9B-ChatSize=9B2024.11 | 23.4 | |
| LoGos(7B)Model Category=Open-source Models (7B), Parameter Count=7B2026.01 | 23.4 | |
| OpenCoder-8B-InstructSize=8B2024.11 | 23.2 | |
| MADAModel=LLaMA3.1-8B, Teacher Model=OpenAI o12026.02 | 23.02 | |
| FLAPBase Model=DeepSeek-R1-Distill-Qwen-7B, Sparsity=12.5%2026.02 | 23 | |
| Star-AgentsModel=LLaMA3.1-8B, Teacher Model=OpenAI o12026.02 | 22.61 | |
| LionModel=Qwen2.5-7B, Teacher Model=OpenAI o12026.02 | 22.34 | |
| CasCoDModel=LLaMA3.1-8B, Teacher Model=OpenAI o12026.02 | 21.94 | |
| CounterDistillModel=LLaMA3.1-8B, Teacher Model=OpenAI o12026.02 | 21.63 | |
| DSSModel=LLaMA3.1-8B, Teacher Model=OpenAI o12026.02 | 21.26 | |
| GeniusTraining Corpus=OpenHermes2025.04 | 21.25 | |
| LaMiniModel=Qwen2.5-14B, Teacher Model=OpenAI o12026.02 | 21.12 | |
| Self-InstructModel=Qwen2.5-14B, Teacher Model=OpenAI o12026.02 | 20.54 | |
| DS-Coder-6.7B-InstructSize=6.7B2024.11 | 20.5 | |
| StarCoder2-15B-Instruct-v0.1Size=15B2024.11 | 20.4 | |
| DeepSeek-R1-Distill-Qwen-7BModel Category=Open-source Models (7B), Parameter Count=7B, Model Variant=Distill2026.01 | 20.4 | |
| LionModel=LLaMA3.1-8B, Teacher Model=OpenAI o12026.02 | 20.16 | |
| CodeQwen1.5-7B-ChatSize=7B2024.11 | 20.1 | |
| LaMiniModel=Qwen2.5-7B, Teacher Model=OpenAI o12026.02 | 20.02 | |
| GeniusTraining Corpus=Magpie2025.04 | 19.75 | |
| LLaMA3.1-8B-Instruct2025.04 | 19.5 | |
| Self-InstructModel=Qwen2.5-7B, Teacher Model=OpenAI o12026.02 | 19.44 | |
| UndistilledModel=Qwen2.5-14B, Teacher Model=OpenAI o12026.02 | 18.42 | |
| LaMiniModel=LLaMA3.1-8B, Teacher Model=OpenAI o12026.02 | 17.92 | |
| UndistilledModel=Qwen2.5-7B, Teacher Model=OpenAI o12026.02 | 17.32 | |
| Self-InstructModel=LLaMA3.1-8B, Teacher Model=OpenAI o12026.02 | 17.25 | |
| Qwen2.5-7B-InstructModel Category=Open-source Models (7B), Parameter Count=7B, Model Variant=Instruct2026.01 | 16.2 | |
| Qwen2.5-Coder-1.5B-InstructSize=1.5B2024.11 | 15.7 | |
| UndistilledModel=LLaMA3.1-8B, Teacher Model=OpenAI o12026.02 | 15.24 | |
| CodeGemma-7B-ItSize=7B2024.11 | 14.7 | |
| Qwen2.5-32B-BaseModel Category=Open-source Models (32B), Parameter Count=32B, Model Variant=Base2026.01 | 13.8 | |
| OpenCoder-1.5B-InstructSize=1.5B2024.11 | 12.8 | |
| Yi-Coder-1.5B-ChatSize=1.5B2024.11 | 11.6 | |
| DS-coder-1.3B-InstructSize=1.3B2024.11 | 9.3 | |
| Qwen2.5-7B-BaseModel Category=Open-source Models (7B), Parameter Count=7B, Model Variant=Base2026.01 | 9 | |
| CrystalChat-7BSize=7B2024.11 | 6.1 | |
| CodeLlama-7B-InstructSize=7B2024.11 | 2.8 |