CommonModelParameters interface
Common language model parameters for Chat Completions. If omitted, default values are used.
Properties
| frequency |
A float in the range [-2,2] that reduces or increases likelihood of repeated tokens. Default is 0. |
| max |
Maximum number of tokens to generate. |
| model | The name of the model to use (e.g., 'gpt-4o', etc.). Default is null if not specified. |
| presence |
A float in the range [-2,2] that penalizes new tokens based on their existing presence. Default is 0. |
| seed | Random seed for controlling deterministic outputs. If omitted, randomization is used. |
| stop | List of stop sequences that will cut off text generation. Default is none. |
| temperature | Sampling temperature. Default is 0.7. |
Property Details
frequencyPenalty
A float in the range [-2,2] that reduces or increases likelihood of repeated tokens. Default is 0.
frequencyPenalty?: number
Property Value
number
maxTokens
Maximum number of tokens to generate.
maxTokens?: number
Property Value
number
model
The name of the model to use (e.g., 'gpt-4o', etc.). Default is null if not specified.
model?: string
Property Value
string
presencePenalty
A float in the range [-2,2] that penalizes new tokens based on their existing presence. Default is 0.
presencePenalty?: number
Property Value
number
seed
Random seed for controlling deterministic outputs. If omitted, randomization is used.
seed?: number
Property Value
number
stop
List of stop sequences that will cut off text generation. Default is none.
stop?: string[]
Property Value
string[]
temperature
Sampling temperature. Default is 0.7.
temperature?: number
Property Value
number