YandexGPT API, REST: TextGenerationAsync.instruct
RPC method for instructing the model to generate text.
HTTP request
POST https://llm.api.cloud.yandex.net/llm/v1alpha/instructAsync
Body parameters
{
"model": "string",
"generationOptions": {
"partialResults": true,
"temperature": "number",
"maxTokens": "integer"
},
// includes only one of the fields `instructionText`, `instructionUri`
"instructionText": "string",
"instructionUri": "string",
// end of the list of possible fields
"requestText": "string"
}
Request for instructing the model to generate text.
Field | Description |
---|---|
model | string The name or identifier of the model to be used for text generation. Possible value for now: The maximum string length in characters is 50. |
generationOptions | object Configuration options for text generation. |
generationOptions. partialResults |
boolean (boolean) Enables streaming of partially generated text. |
generationOptions. temperature |
number (double) Affects creativity and randomness of responses. Should be a double number between 0 (inclusive) and 1 (inclusive). Lower values produce more straightforward responses, while higher values lead to increased creativity and randomness. |
generationOptions. maxTokens |
integer (int64) Sets the maximum limit on the total number of tokens used for both the input prompt and the generated response. Must be greater than zero and not exceed 7400 tokens. |
instructionText | string includes only one of the fields instructionText , instructionUri The text-based instruction for text generation. |
instructionUri | string includes only one of the fields instructionText , instructionUri A URI containing instructions for text generation. |
requestText | string The text-based request for text generation. |
Response
HTTP Code: 200 - OK
{
"id": "string",
"description": "string",
"createdAt": "string",
"createdBy": "string",
"modifiedAt": "string",
"done": true,
"metadata": "object",
// includes only one of the fields `error`, `response`
"error": {
"code": "integer",
"message": "string",
"details": [
"object"
]
},
"response": "object",
// end of the list of possible fields
}
An Operation resource. For more information, see Operation.
Field | Description |
---|---|
id | string ID of the operation. |
description | string Description of the operation. 0-256 characters long. |
createdAt | string (date-time) Creation timestamp. String in RFC3339 text format. The range of possible values is from To work with values in this field, use the APIs described in the Protocol Buffers reference. In some languages, built-in datetime utilities do not support nanosecond precision (9 digits). |
createdBy | string ID of the user or service account who initiated the operation. |
modifiedAt | string (date-time) The time when the Operation resource was last modified. String in RFC3339 text format. The range of possible values is from To work with values in this field, use the APIs described in the Protocol Buffers reference. In some languages, built-in datetime utilities do not support nanosecond precision (9 digits). |
done | boolean (boolean) If the value is |
metadata | object Service-specific metadata associated with the operation. It typically contains the ID of the target resource that the operation is performed on. Any method that returns a long-running operation should document the metadata type, if any. |
error | object The error result of the operation in case of failure or cancellation. includes only one of the fields error , response |
error. code |
integer (int32) Error code. An enum value of google.rpc.Code. |
error. message |
string An error message. |
error. details[] |
object A list of messages that carry the error details. |
response | object includes only one of the fields error , response The normal response of the operation in case of success. If the original method returns no data on success, such as Delete, the response is google.protobuf.Empty. If the original method is the standard Create/Update, the response should be the target resource of the operation. Any method that returns a long-running operation should document the response type, if any. |