Syrus AI Chat Model

Chat Model Syrus AI (Custom Implementation)

AI-generated

Summary

Generate a text completion from the Syrus AI chat model by providing parameters controlling sampling and output length.

Temperature — Controls randomness in token sampling, higher means more diverse outputs (default 0.7).
Top P — Nucleus sampling threshold limiting tokens to those accumulating to probability Top P (default 0.95).
Top K — Limits tokens considered to the top K highest probability tokens during sampling (default 40).
Max Output Tokens — Maximum number of tokens to generate in the completion (default 8192).

an object containing an array 'generations' with one generation entry that includes the generated text and AIMessage object

The output includes the first candidate text generated by the Syrus AI model. Errors are thrown on API failure.