Reward Modeling on VisionArena Battle
86.25AccuracyIXC-2.5-Reward
Evaluation Results
| Method | Links | |
|---|---|---|
| IXC-2.5-RewardModel Category=Open-source Specialist RMs2026.04 | 86.25 | |
| Claude 4.6 OpusModel Category=Proprietary Frontier VLMs2026.04 | 78.83 | |
| Qwen3-VL-32B-InstructModel Category=Open-source Generalist VLMs, Prompting Strategy=Rubric2026.04 | 76.83 | |
| Gemini 3.0 FlashModel Category=Proprietary Frontier VLMs2026.04 | 76.66 | |
| Gemini 3.1 ProModel Category=Proprietary Frontier VLMs2026.04 | 76.25 | |
| Qwen3-VL-30B-A3B-InstructModel Category=Open-source Generalist VLMs, Prompting Strategy=Rubric2026.04 | 75.5 | |
| Qwen3-VL-32B-InstructModel Category=Open-source Generalist VLMs, Prompting Strategy=Vanilla2026.04 | 75.42 | |
| Qwen3-VL-32B-InstructModel Category=Open-source Generalist VLMs, Prompting Strategy=CoT2026.04 | 74.58 | |
| Gemini 3.1 Flash-LiteModel Category=Proprietary Frontier VLMs2026.04 | 74.33 | |
| GPT-5.4Model Category=Proprietary Frontier VLMs2026.04 | 74.25 | |
| Qwen3-VL-30B-A3B-InstructModel Category=Open-source Generalist VLMs, Prompting Strategy=Vanilla2026.04 | 74.08 | |
| Qwen3-VL-30B-A3B-InstructModel Category=Open-source Generalist VLMs, Prompting Strategy=CoT2026.04 | 72.17 | |
| Skywork-VL RewardModel Category=Open-source Specialist RMs2026.04 | 71.67 |