LlamaFarm provides intuitive command line tools to operate the RAG system, consisting of two key processes:
1. Data entry (Ingest)::
- Prepare the folder containing the documents (e.g., samples/)
- Execute the command:
uv run python rag/cli.py ingest samples/ --extractors keywords entities --strategy research - This command will: automatically recognize the document format → extract keywords/entities → chunk by RESEARCH strategy → store to default vector database
2. Knowledge retrieval (Search)::
- Example of query command:
uv run python rag/cli.py search "气候变暖的主要证据" --top-k 5 --rerank - Parameter Description:
- -top-k 5: return the 5 most relevant fragments
- -rerank: Enabling reordering of results improves accuracy
- The system will first retrieve the vector library and then optimize the results by reordering the model
The entire process requires no coding and can be adapted to different document types by adjusting the policy parameters.
This answer comes from the articleLlamaFarm: a development framework for rapid local deployment of AI models and applicationsThe
May not be reproduced without permission:AI productivity tools " How can I use LlamaFarm for knowledge base creation and retrieval via command line? Please explain the exact procedure
English 
简体中文 
日本語 
Deutsch 
Português do Brasil