Usage

friendli api chat-completions create [OPTIONS]

Summary

Creates chat completions.

Options

OptionTypeSummaryDefaultRequired
--message, -gTEXTA message in ROLE CONTENT format. Repeat this option to add multiple messages.-
--model, -mTEXTThe model to use for chat completions. See here for more about available models and pricing.-
--n, -nINTEGER RANGEThe number of results to generate.None
--max-tokens, -MINTEGER RANGEThe maximum number of tokens to generate.None
--stop, -STEXTWhen one of the stop phrases appears in the generation result, the API will stop generation. The stop phrases are excluded from the result. Repeat this option to use multiple stop phrases.None
--temperature, -TFLOAT RANGESampling temperature. non-zero positive numbers are allowed.None
--top-p, -PFLOAT RANGETokens comprising the top top_p probability mass are kept for sampling.None
--frequency-penalty, -fpFLOAT RANGEPositive values penalizes tokens that have been sampled, taking into account their frequency in the preceding text. This penalization diminishes the model’s tendency to reproduce identical lines verbatim.None
--presence-penalty, -ppFLOAT RANGEPositive values penalizes tokens that have been sampled at least once in the existing text.None
--stream, -sBOOLEANWhether to stream generation result.False
--token, -tTEXTFriendli Token for auth.None
--team-idTEXTID of team to run as.None