Kevin Hu
Kevin Hu
That happens when deepseek responds abnormal things.
It needs download models from huggingface. You can download them manually and put them into rag/res/deepdoc/.
> huggingface下载什么模型? [Here](https://huggingface.co/InfiniFlow/deepdoc/tree/main).
What about restarting all the containers? When/Where did that happen?
No matter where Ollama run, as long as the network is accessible, RAGFlow can utilize it anyway.
demo.ragflow.io and `host.docker.internal:11434` are not pingable.
There is a parameter called 'cite', set it 'true'.
Sorry, my bad. It can not support cites yet.
When/where does that happen?
I don't think KG is a good choise. Could you make an example why API instructions can't associate with feature documents, and why human being can related them?