multi-mediawiki-rag
multi-mediawiki-rag copied to clipboard
Prompt Compression
Is your feature request related to a problem? Please describe. The context given from a retrieval can be compressed before given to Ollama for processing, which allows for more text to be given as context and better performance.
Describe the solution you'd like Langchain Contextual Compression.
Describe alternatives you've considered Prompt Compression.