Kevin Hu
Kevin Hu
Documentation is a good suggestion. No influence to local volumes by upgrading system. If it will, we will claim that in release note. BTW: Never **`docker compose down -v`** which...
It's probably caused by abnormal response from LLMs.
Which embedding models did you use?
Is [this](https://github.com/infiniflow/ragflow/issues/531#issuecomment-2611476179) helpful?
This largly depends on the data quality of DDL, DB description and Q-SQL. LLM generates SQL by the content retrieved from these 3 knowledge base as samples and clues. Upgrade...
The tag `latest` has been deprecated.
Pull the `nightly` version of docker image.
1. If it's deployed without SAAS invokation, you need to deploy a multimodual LLM locally. RAGFlow can invoke them by OpenAI compatible API/Ollama/Xinference. 2. Check out the template 'Research report...
Did you apply `Laws/Book` as the chunking method? They generate overlaped chunks. Try `General`.
Could you paste the backend error message here? `docker logs -f ragflow-server`