infiniflow / ragflow

RAGFlow is an open-source RAG (Retrieval-Augmented Generation) engine based on deep document understanding.
https://ragflow.io
Apache License 2.0
18.18k stars 1.84k forks source link

[Question]: Qwen2-72B-Instruct-GPTQ-Int4 of Xinference not listed in System model settings #2506

Open 0000sir opened 1 week ago

0000sir commented 1 week ago

Describe your problem

I'm running Qwen2-72B-Instruct-GPTQ-Int4 with Xinference, after add model to ragflow, I can't select it from the Chat model drop down list, it's not available in this list. I read the codes but didn't find any reason, but a embedding model bge-large-zh-v1.5 is listed in Embedding model selection.

What should I do to make it work, thanks.

image

As you can see the Qwen2 model is shown under the overlay below Xinference, but not listed in the selection.