List your company
ResearchFormerly WizWand

Multimodal Understanding on SEED-Bench Image (test)

75.9AccuracyInternLM-XComposer2
23.48437.09250.764.308Nov 21, 2023Dec 2, 2023Dec 14, 2023Dec 25, 2023Jan 6, 2024Jan 17, 2024Jan 29, 2024
Updated 8mo ago

Evaluation results

MethodLinks
InternLM-XComposer2LLM=InternLM2-7B2024.0175.9
LLaVA-XTunerLLM=InternLM2-20B2024.0170.2
ShareGPT4V-7BLanguage Model=Vicuna-7B2023.1169.7
ShareGPT4VLLM=Vicuna-7B2024.0169.7
Monkey-10BLLM=Qwen-7B2024.0168.9
CogVLM-17BLLM=Vicuna-7B2024.0168.8
LLaVA-1.5Language Model=Vicuna-13B2023.1168.2
LLaVA-1.5LLM=Vicuna-13B2024.0168.2
LLaVA-1.5Language Model=Vicuna-7B2023.1166.2
InternLM-XCLLM=InternLM-7B2024.0166.1
Qwen-VL-ChatLanguage Model=Qwen-7B2023.1158.2
Qwen-VL-ChatLLM=Qwen-7B2024.0158.2
Qwen-VLLanguage Model=Qwen-7B2023.1156.3
InstructBLIPLanguage Model=Vicuna-7B2023.1153.4
InstructBLIPLLM=Vicuna-7B2024.0153.4
IDEFICS-80BLLM=LLaMA-65B2024.0152
BLIP-2Language Model=FLAN-T52023.1146.4
BLIP-2LLM=FLAN-T52024.0146.4
LLaVALanguage Model=Vicuna-7B2023.1125.5
LLaVALLM=Vicuna-7B2024.0125.5