Visual Question Answering on V2X-QA vehicle-side 1.0 (test)
98.9Accuracy (VS1)V2X-MoE-8B
Evaluation Results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| V2X-MoE-8BModel Type=Open-Source2026.04 | 98.9 | 94.3 | 93.2 | 94.7 | 95.3 | |
| Gemini-2.5-ProModel Type=Closed-Source2026.04 | 87.2 | 59.2 | 40.4 | 46.8 | 58.4 | |
| Qwen-3.5-PlusModel Type=Closed-Source2026.04 | 84.5 | 55.1 | 44.5 | 58.9 | 60.8 | |
| Intern-S1Model Type=Open-Source2026.04 | 83.8 | 63.8 | 42.3 | 44.2 | 58.5 | |
| GPT-5-MiniModel Type=Closed-Source2026.04 | 78.1 | 49.8 | 32.1 | 44.5 | 51.1 | |
| Qwen-3-VL-30B-A3B-InstructModel Type=Open-Source2026.04 | 72.1 | 44.2 | 27.5 | 43.4 | 46.8 | |
| Qwen-2.5-VL-72B-InstructModel Type=Open-Source2026.04 | 72.1 | 65.3 | 51.3 | 41.9 | 57.7 | |
| Gemini-3-Flash-PreviewModel Type=Closed-Source2026.04 | 69.4 | 56.2 | 20.8 | 38.9 | 46.3 | |
| Qwen3-VL-FlashModel Type=Closed-Source2026.04 | 65.5 | 49.5 | 39 | 55.8 | 52.5 | |
| GPT-5.2Model Type=Closed-Source2026.04 | 65.3 | 43.8 | 26 | 43 | 44.5 | |
| Intern-S1-MiniModel Type=Open-Source2026.04 | 47.2 | 52.1 | 29.8 | 43.4 | 43.1 |