Controls the maximum number of tokens to generate.

Controls the model's tendency to generate repetitive responses. Lower values make the model more likely to repeat information.

Controls the maximum number of tokens to generate and in turn the response length.

Controls the avoidance of specific topics in its responses. Lower values make the model less concerned about preventing those topics.

Controls the model's tendency to generate repetitive responses. Values > 1 penalize repetitions.

Change the seed to randomize responses if your getting the same output over and over. Does not guarantee deterministic outputs.

Controls randomness. Lower values make the output more focused and deterministic.

Limits the sample to the top K most likely tokens. Lower values make the model more deterministic.

Also known as nucleus sampling or 'penalty', this controls the diversity and quality of the responses. It limits the cumulative probability of the most likely tokens. Lower values lead to more constrained responses.

A string or a list of strings that, when encountered in the generated response, will cause the model to stop generating further tokens.