internlm

/

internlm2_5-7b-chat

like 89

There are a number of tool call issues with the version of the model in this repo. There are also a various of formatting inconsistencies, especially between the slow and fast tokenizers.

I have fixed the issues and:

• Posted a new copy of the model.
• Created llama.cpp/GGUF quants.

Both are here on HF.

Thanks!