Multimodal Understanding on NaturalBench
77.2NaturalBench ScoreLLaVA-OneVision
Evaluation results
| Method | Links | |
|---|---|---|
| LLaVA-OneVisionLLM=Qwen2-7B2025.05 | 77.2 | |
| LLaVA-NeXTLLM=Vicuna-7B2025.05 | 70.2 | |
| Slot-MLLM (14B)LLM=Qwen2.5-14B-Instruct, PT Data Size=25M2025.05 | 69.8 | |
| Slot-MLLM (14B)LLM=Qwen2.5-14B-Instruct2025.05 | 69.8 | |
| ShareGPT4VLLM=Vicuna-7B2025.05 | 68.4 | |
| Slot-MLLM (7B)LLM=Vicuna-7B, PT Data Size=25M2025.05 | 66.1 | |
| Slot-MLLM (7B)LLM=Vicuna-7B2025.05 | 66.1 | |
| VILA-ULLM=LLaMA-2-7B, PT Data Size=23M2025.05 | 63.1 | |
| InstructBLIPLLM=Vicuna-7B2025.05 | 59.1 | |
| SEED-LLaMALLM=Vicuna-7B, PT Data Size=77M2025.05 | 57.5 | |
| LaVITLLM=LLaMA-7B, PT Data Size=100M2025.05 | 52.5 | |
| UniTokLLM=LLaMA-2-7B, PT Data Size=70M2025.05 | 50.6 |