AddCustomModelRequestBody

Configuration for registering a custom model. Defines the model's capabilities, endpoint, and default parameters.

Example Usage

import { AddCustomModelRequestBody } from "@meetkai/mka1/models/operations";

let value: AddCustomModelRequestBody = {
  name: "<value>",
  baseUrl: "https://definitive-responsibility.info/",
};

Fields

Field	Type	Required	Description
name	string	✔️	Unique name for the custom model. Must be between 1 and 255 characters. This name is used to reference the model in API requests.
baseUrl	string	✔️	Base URL of the custom model API endpoint. Must be a valid URL (maximum 512 characters). This is where API requests for this model will be sent.
apiKey	string	➖	Optional API key for authenticating with the custom model endpoint. Maximum 512 characters. Include if the custom model requires authentication.
supportsChat	boolean	➖	Whether this model supports chat completions (conversational interactions). Set to true if the model can handle multi-turn conversations.
supportsCompletion	boolean	➖	Whether this model supports text completions (single-turn text generation). Set to true if the model can generate text from a prompt.
supportsEmbedding	boolean	➖	Whether this model supports text embeddings (vector representations). Set to true if the model can generate semantic embeddings.
supportsImage	boolean	➖	Whether this model supports image generation or image understanding. Set to true if the model can work with images.
supportsTranscription	boolean	➖	Whether this model supports audio transcription (speech-to-text). Set to true if the model can transcribe audio.
supportsSpeech	boolean	➖	Whether this model supports speech synthesis (text-to-speech). Set to true if the model can generate audio from text.
maxTokens	number	➖	Default maximum number of tokens the model can generate in a single response. Must be at least 1. This limits the length of generated outputs.
contextWindow	number	➖	The context window size of the model (total tokens including prompt and completion). Must be at least 1. This determines how much text the model can process at once.
temperature	number	➖	Default sampling temperature for the model (0-2). Higher values (e.g., 0.8) make output more random, lower values (e.g., 0.2) make it more deterministic.
topP	number	➖	Default nucleus sampling parameter (0-1). Alternative to temperature. Only tokens with cumulative probability up to topP are considered.
frequencyPenalty	number	➖	Default frequency penalty (-2 to 2). Positive values reduce repetition of tokens based on their frequency in the generated text.
presencePenalty	number	➖	Default presence penalty (-2 to 2). Positive values encourage the model to talk about new topics by penalizing tokens that have appeared.
rpm	number	➖	Rate limit in requests per minute for this model. Must be at least 1. Used for throttling requests to prevent overloading the custom model endpoint.