Multimodal Reasoning on VRC-Bench 1.0 (test)
61.35Final ScoreClaude-3.5 Sonnet
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| Claude-3.5 SonnetSource Type=Close-Source2025.01 | 61.35 | 72.12 | |
| Gemini-1.5 ProSource Type=Close-Source2025.01 | 61.35 | 72.12 | |
| Gemini-2.0 FlashSource Type=Close-Source2025.01 | 61.16 | 74.08 | |
| GPT-4oSource Type=Close-Source2025.01 | 59.28 | 76.68 | |
| LlamaV-01Source Type=Open-Source2025.01 | 56.49 | 68.93 | |
| GPT-4o miniSource Type=Close-Source2025.01 | 56.39 | 74.05 | |
| Gemini-1.5 FlashSource Type=Close-Source2025.01 | 54.99 | 71.86 | |
| Llava-CoTSource Type=Open-Source2025.01 | 54.09 | 66.21 | |
| MulberrySource Type=Open-Source2025.01 | 51.9 | 63.86 | |
| Llama-3.2 VisionSource Type=Open-Source2025.01 | 48.4 | 58.37 |