General Multimodal Evaluation on LLaVA-Bench Wild
92.8Relative ScoreTroL-7B
Evaluation results
| Method | Links | |
|---|---|---|
| TroL-7BParameters=7B, Evaluator=GPT-4-06132024.06 | 92.8 | |
| LLaVA-NeXT-8BParameters=8B, Evaluator=GPT-4-06132024.06 | 80.1 | |
| CSRModel Scale=13B, Iterations=iter-32024.05 | 74.7 | |
| LLaVA-NeXT-7BParameters=7B, Evaluator=GPT-4-06132024.06 | 72.3 | |
| LLaVA-NeXT-13BParameters=13B, Evaluator=GPT-4-06132024.06 | 72.3 | |
| CSRModel Scale=7B, Iterations=iter-32024.05 | 71.1 | |
| Qwen-VL-Chat2024.05 | 67.7 | |
| LLaVA1.5-7BParameters=7B, Evaluator=GPT-4-06132024.06 | 63.4 | |
| InstructBLIP2024.05 | 60.9 | |
| mPLUG-Owl22024.05 | 59.9 | |
| IDEFICS2024.05 | 45 | |
| BLIP-22024.05 | 38.1 |