Instruction Following Evaluation on ArenaHard v1
38ArenaHardv1 Score+RL (Skywork-Reward-V2-Llama-3.1-8B)
Evaluation Results
| Method | Links | |
|---|---|---|
| +RL (Skywork-Reward-V2-Llama-3.1-8B)Model=Qwen2.5-7B2025.07 | 38 | |
| Instruct (official)Model=Qwen2.5-7B2025.07 | 37.9 | |
| +RL (Skywork-Reward-V2-Qwen3-4B)Model=Qwen2.5-7B2025.07 | 35 | |
| +RL (Skywork-Reward-Gemma-2-27B-v0.2)Model=Qwen2.5-7B2025.07 | 34.5 | |
| +RL (Skywork-Reward-Llama-3-8B-v0.2)Model=Qwen2.5-7B2025.07 | 29.8 | |
| Instruct (official)Model=Llama-3.1-8B2025.07 | 24.9 | |
| +SFTModel=Qwen2.5-7B2025.07 | 22.1 | |
| +RL (Skywork-Reward-V2-Llama-3.1-8B)Model=Llama-3.1-8B2025.07 | 20.8 | |
| +RL (Skywork-Reward-V2-Qwen3-4B)Model=Llama-3.1-8B2025.07 | 18.8 | |
| BaseModel=Qwen2.5-7B2025.07 | 16.2 | |
| +RL (Skywork-Reward-Gemma-2-27B-v0.2)Model=Llama-3.1-8B2025.07 | 14 | |
| +SFTModel=Llama-3.1-8B2025.07 | 12.6 | |
| +RL (Skywork-Reward-Llama-3-8B-v0.2)Model=Llama-3.1-8B2025.07 | 9.7 | |
| BaseModel=Llama-3.1-8B2025.07 | 6.8 |