archie0729/llava-en-zh-300k

This dataset is composed by 150k examples of English Visual Instruction Data from LLaVA. 150k examples of English Visual Instruction Data from openbmb. You can use it in LLaMA Factory by specifying --dataset llava_150k_en,llava_150k_zh.