Optional
temperatureControls randomness in output (0.0 to 2.0)
Optional
top_Controls nucleus sampling (0.0 to 1.0)
Optional
nNumber of completions to generate
Optional
stopSequences where the API will stop generating
Optional
presence_Penalizes new tokens based on their presence in the text so far
Optional
frequency_Penalizes new tokens based on their frequency in the text so far
Optional
seedSeed for deterministic generation
Optional
max_Maximum number of tokens to generate
Optional
logit_Modify the likelihood of specified tokens appearing
Optional
userUser identifier for tracking and abuse monitoring
Extended OpenAI parameters interface for fine-tuning model behavior. Provides comprehensive control over model generation parameters.