internlm/internlm-xcomposer2d5-7b model request (multimodal)

[internlm/internlm-xcomposer2d5-7b](https://huggingface.co/internlm/internlm-xcomposer2d5-7b) > InternLM-XComposer2.5 excels in various text-image comprehension and composition applications, achieving GPT-4V level capabilities with merely 7B LLM backend. IXC2.5 is trained with 24K interleaved image-text contexts, it can seamlessly extend to 96K long contexts via RoPE extrapolation. This long-context capability allows IXC-2.5 to excel in tasks requiring extensive input and output contexts. https://github.com/InternLM/InternLM-XComposer