Skip to main contentHere you’ll find the main parameters used to configure RAG agents on the Contextual AI platform. Each parameter includes a sensible default, which you can modify to optimize performance for your specific data and query patterns.
This reference is organized into the following sections:
Core configuration options that apply to Contextual AI agents, such as model selection, default behaviors, and key runtime settings. Use these parameters as your baseline before layering on more advanced controls.
Instruct the agent on how to respond to users’ queries given the retrieved knowledge. The appropriate prompt is passed, along with the user query and relevant retrievals, to the Generator Model at generation-time.
These settings affect if and how user queries are modified to improve retrieval performance and response generation.
These settings allow you to modify the original user query prior to retrieval and generation. These strategies can help improve retrieval accuracy or completeness.
These settings determine how the agent performs the initial retrieval from linked unstructured datastores.
These settings affect how the agent reranks and filters chunks before passing them to the generator model.
These settings affect how the generator model produces responses.
Additional controls for refining edge cases and improving overall user experience. 
