68e7c297e0
- Add instructions for self-hosted users with info, warning boxes to
avoid, fix common issues when setting up Khoj server
- Create new Advanced Self Hosting section
- Extract Advanced Self-Hosting Sections from the Advanced Page and
move them to separate Pages under Advanced Self Hosting section
- Improve OpenAI Proxy Docs
- Put Ollama setup as a section under OpenAI API Proxy page instead
of a separate page
- Add Section to use Khoj with chat model from LM Studio
- Update LiteLLM docs to use chat model from LM Studio
1.9 KiB
Support Multilingual Docs
Khoj uses an embedding model to understand documents. Multilingual embedding models improve the search quality for documents not in English. This affects both search and chat with docs experiences across Khoj.
To improve search and chat quality for non-english documents you can use a multilingual model.
For example, the paraphrase-multilingual-MiniLM-L12-v2 supports 50+ languages, has decent search quality and speed for a consumer machine.
To use it:
- Open the search config on your server's admin settings page. Either create a new search model, if none exists, or update the existing one. For example,
- Set the
bi_encoderfield tosentence-transformers/paraphrase-multilingual-MiniLM-L12-v2 - Set the
cross_encoderfield tomixedbread-ai/mxbai-rerank-xsmall-v1
- Set the
- Regenerate your content index from all the relevant clients. This step is very important, as you'll need to re-encode all your content with the new model.
:::info[Note]
Modern search/embedding model like mixedbread-ai/mxbai-embed-large-v1 expect a prefix to the query (or docs) string to improve encoding. Update the bi_encoder_query_encode_config field of your embedding model with {prompt: <prefix-prompt>} to improve the search quality of these models.
E.g. {prompt: "Represent this query for searching documents"}. You can pass any valid JSON object that the SentenceTransformer encode function accepts
:::