Interact with the model. Select your model, set your system prompt, and adjust parameters on the left.
Choose from available models.
Include recent search context to improve answers.
Define the base context for the AI's responses.
Maximum tokens for the response.
Controls the randomness of the output.
Limits token candidates to the top-k tokens.
Limits token candidates to a cumulative probability threshold.
Penalizes token repetition to improve diversity.
Maximum number of search results to retrieve.
Maximum characters to retrieve per search result.