Image Quality Assessment on QBench (test)
74.1AccuracyGPT-4V
Evaluation results
| Method | Links | ||
|---|---|---|---|
| GPT-4V2024.01 | 74.1 | — | |
| InternLM-XComposer2LLM=InternLM2-7B2024.01 | 72.5 | — | |
| InternLM-XComposer2Number of Parameters=7B2024.01 | 72.5 | — | |
| Gemini-Pro2024.01 | 70.6 | — | |
| Qwen-VL-Plus2024.01 | 68.9 | — | |
| InternLM-XCLLM=InternLM-7B2024.01 | 64.4 | — | |
| Int-XCNumber of Parameters=8B2024.01 | 64.4 | — | |
| ShareGPT4V-7BLanguage Model=Vicuna-7B2023.11 | 63.4 | — | |
| LLaVA-1.5Language Model=Vicuna-13B2023.11 | 62.1 | — | |
| Qwen-VL-ChatLLM=Qwen-7B2024.01 | 61.7 | — | |
| LLaVA-1.5LLM=Vicuna-13B2024.01 | 61.4 | — | |
| Qwen-VLLanguage Model=Qwen-7B2023.11 | 59.4 | — | |
| LLaVA-1.5Language Model=Vicuna-7B2023.11 | 58.7 | — | |
| InstructBLIPLanguage Model=Vicuna-7B2023.11 | 56.7 | — | |
| InstructBLIPLLM=Vicuna-7B2024.01 | 55.9 | — | |
| ShikraLanguage Model=Vicuna-13B2023.11 | 54.7 | — | |
| LLaVALLM=Vicuna-7B2024.01 | 54.7 | — | |
| InstructBLIPLLM Backbone=Vicuna-7B2023.12 | — | 55.9 | |
| InternLM-XCLLM=InernLM-7B2024.04 | — | 64.4 | |
| IXC2-4KHDLLM=InernLM2-7B2024.04 | — | 71.8 | |
| IXC2-VLLLM=InernLM2-7B2024.04 | — | 72.5 | |
| LLaVALLM Backbone=Vicuna-7B2023.12 | — | 54.7 | |
| LLaVA-1.5LLM=Vicuna-13B2024.04 | — | 61.4 | |
| LLaVA-1.5LLM Backbone=Vicuna-13B2023.12 | — | 61.4 | |
| LyricsLLM Backbone=Vicuna-13B2023.12 | — | 72.5 | |
| Qwen-VL-ChatLLM=Qwen-7B2024.04 | — | 61.7 | |
| Qwen-VL-ChatLLM Backbone=Qwen-7B2023.12 | — | 61.7 |