> ## Documentation Index > Fetch the complete documentation index at: https://docs.creatify.ai/llms.txt > Use this file to discover all available pages before exploring further. # Create Text Generator task > High-performance text generation endpoint on the async Starlette service. Authenticates via API token (X-API-KEY + X-API-ID). Supports two modes: - **Sync** (`sync: true`): blocks until generation completes, returns the result directly. - **Async** (default): returns immediately with `pending` status. Poll `GET /api/text_generator/{id}/` or use a webhook. ## OpenAPI ````yaml post /sse/text_generator/ openapi: 3.0.3 info: title: creatify.ai API version: 1.0.0 description: API for creatify.ai servers: [] security: [] paths: /sse/text_generator/: post: tags: - text_generator summary: Create Text Generator task description: >- High-performance text generation endpoint on the async Starlette service. Authenticates via API token (X-API-KEY + X-API-ID). Supports two modes: - **Sync** (`sync: true`): blocks until generation completes, returns the result directly. - **Async** (default): returns immediately with `pending` status. Poll `GET /api/text_generator/{id}/` or use a webhook. operationId: sse_text_generator_create requestBody: required: true content: application/json: schema: $ref: '#/components/schemas/TextGenRequest' responses: '201': description: Text generation result or pending task. content: application/json: schema: $ref: '#/components/schemas/TextGenResponse' '400': description: Invalid request body or model not found. content: application/json: schema: type: object properties: error: type: string '401': description: Missing or invalid API credentials. '402': description: Not enough credits. '429': description: Rate limit exceeded. security: - X-API-ID: [] X-API-KEY: [] components: schemas: TextGenRequest: type: object properties: model_name: type: string description: Name of the Gemini model (e.g. gemini-2.5-flash). maxLength: 255 messages: type: array items: $ref: '#/components/schemas/Message' description: List of messages with 'role' (user/model) and 'content'. system_instruction: type: string default: '' description: Optional system instruction for the model. config: allOf: - $ref: '#/components/schemas/TextGenConfig' description: Generation config parameters. tools: type: array items: $ref: '#/components/schemas/Tool' description: List of tools (function declarations) the model may call. tool_config: allOf: - $ref: '#/components/schemas/ToolConfig' description: Controls how the model uses the provided tools. webhook_url: type: string format: uri description: Webhook URL for async status updates. sync: type: boolean default: false description: >- If true, the request blocks until generation completes and returns the result directly. Default is false (async). required: - messages - model_name TextGenResponse: type: object properties: id: type: string format: uuid readOnly: true model_name: type: string readOnly: true messages: type: array items: $ref: '#/components/schemas/Message' readOnly: true system_instruction: type: string description: Optional system instruction for the model. config: allOf: - $ref: '#/components/schemas/TextGenConfig' readOnly: true status: $ref: '#/components/schemas/TextGenResponseStatusEnum' response_text: type: string description: Generated text response from the model. response_function_calls: type: array items: $ref: '#/components/schemas/FunctionCallPart' readOnly: true failed_reason: type: string description: Reason for failure if status is failed. credits_used: type: number format: double description: Credits consumed for this generation. created_at: type: string format: date-time readOnly: true title: Created date updated_at: type: string format: date-time readOnly: true title: Updated date required: - config - created_at - id - messages - model_name - response_function_calls - updated_at Message: type: object properties: role: allOf: - $ref: '#/components/schemas/RoleEnum' description: |- Role of the message sender: 'user' or 'model'. * `user` - user * `model` - model content: default: '' description: >- Message content. Either a plain text string, or a list of content parts for multimodal input. Each part is an object with 'type' ('text', 'image', or 'video'), and either 'text' (for text parts) or 'data' (base64) + 'mime_type' (for image/video parts). function_call: allOf: - $ref: '#/components/schemas/FunctionCallPart' description: A function call part (role must be 'model'). function_response: allOf: - $ref: '#/components/schemas/FunctionResponsePart' description: A function response part (role must be 'user'). required: - role TextGenConfig: type: object description: Structured config serializer for OpenAPI documentation. properties: temperature: type: number format: double description: 'Controls randomness. Range: 0.0 - 2.0. Lower = more deterministic.' max_output_tokens: type: integer description: Maximum number of tokens to generate. top_p: type: number format: double description: 'Nucleus sampling threshold. Range: 0.0 - 1.0.' top_k: type: integer description: Top-k sampling. Only sample from top k tokens. presence_penalty: type: number format: double description: 'Penalize tokens already present in the text. Range: -2.0 - 2.0.' frequency_penalty: type: number format: double description: 'Penalize tokens by frequency. Range: -2.0 - 2.0.' seed: type: integer description: Random seed for deterministic generation. stop_sequences: type: array items: type: string description: List of strings that stop generation when encountered. response_mime_type: allOf: - $ref: '#/components/schemas/ResponseMimeTypeEnum' description: |- Output format: 'text/plain' or 'application/json'. * `text/plain` - text/plain * `application/json` - application/json response_json_schema: description: >- JSON Schema for structured output. Automatically sets response_mime_type to 'application/json'. Tool: type: object description: A tool containing one or more function declarations. properties: function_declarations: type: array items: $ref: '#/components/schemas/FunctionDeclaration' description: List of function declarations available to the model. required: - function_declarations ToolConfig: type: object description: Controls how the model uses the provided tools. properties: function_calling_config: allOf: - $ref: '#/components/schemas/FunctionCallingConfig' description: Function calling configuration. TextGenResponseStatusEnum: enum: - pending - running - done - failed type: string description: |- * `pending` - pending * `running` - running * `done` - done * `failed` - failed FunctionCallPart: type: object description: A function call issued by the model. properties: name: type: string description: Name of the function to call. args: type: object additionalProperties: {} description: Arguments for the function call as a JSON object. required: - name RoleEnum: enum: - user - model type: string description: |- * `user` - user * `model` - model FunctionResponsePart: type: object description: A function response supplied by the caller. properties: name: type: string description: Name of the function that was called. response: type: object additionalProperties: {} description: The function's return value as a JSON object. required: - name - response ResponseMimeTypeEnum: enum: - text/plain - application/json type: string description: |- * `text/plain` - text/plain * `application/json` - application/json FunctionDeclaration: type: object description: Declares a function the model may call. properties: name: type: string description: Function name. description: type: string default: '' description: Description of what the function does. parameters: description: JSON Schema describing the function's parameters. required: - name FunctionCallingConfig: type: object description: Controls how the model selects and calls functions. properties: mode: allOf: - $ref: '#/components/schemas/ModeEnum' description: |- Function calling mode: AUTO (default), ANY, NONE, or VALIDATED. * `AUTO` - AUTO * `ANY` - ANY * `NONE` - NONE * `VALIDATED` - VALIDATED allowed_function_names: type: array items: type: string description: >- Restrict the model to only call these functions (used with mode=ANY). ModeEnum: enum: - AUTO - ANY - NONE - VALIDATED type: string description: |- * `AUTO` - AUTO * `ANY` - ANY * `NONE` - NONE * `VALIDATED` - VALIDATED securitySchemes: X-API-ID: type: apiKey in: header name: X-API-ID description: API ID, from your settings page. X-API-KEY: type: apiKey in: header name: X-API-KEY description: API Key, from your settings page. ````