> ## Documentation Index > Fetch the complete documentation index at: https://docs.subconscious.dev/llms.txt > Use this file to discover all available pages before exploring further. # Create Response > Generates a model response from an input sequence. This endpoint is compatible with the OpenAI Responses API, which Codex and the Vercel AI SDK use. Session affinity works the same as on chat completions: a body session id, a session header, or prompt_cache_key groups requests into one conversation. ## OpenAPI ````yaml api-reference/openapi.json POST /responses openapi: 3.1.0 info: title: Subconscious API version: 1.0.0 description: OpenAI-compatible inference API servers: - url: https://api.subconscious.dev/v1 security: - bearerAuth: [] paths: /responses: post: summary: Create a response description: >- Generates a model response from an input sequence. This endpoint is compatible with the OpenAI Responses API, which Codex and the Vercel AI SDK use. Session affinity works the same as on chat completions: a body session id, a session header, or prompt_cache_key groups requests into one conversation. operationId: createResponse parameters: - name: x-subconscious-client in: header required: false description: >- Labels your traffic in usage views. See [Request correlation](https://subconscious.dev/features/request-correlation). schema: type: string - name: x-yourapp-session-id in: header required: false description: >- Groups requests into one conversation (replace `yourapp` with your app name). See [Session affinity](https://subconscious.dev/features/session-affinity). schema: type: string - name: x-yourapp-parent-session-id in: header required: false description: >- Nests a subagent under the conversation that spawned it. See [Subagent correlation](https://subconscious.dev/features/subagent-correlation). schema: type: string requestBody: required: true content: application/json: schema: type: object required: - model - input properties: model: type: string description: The model to use. enum: - subconscious/glm-5.3-marathon example: subconscious/glm-5.3-marathon input: description: >- The input to the model. A string, or an array of input items in the OpenAI Responses format, such as messages, function calls, and function call outputs. example: Write a haiku about GPUs. instructions: type: string description: System instructions for the model. tools: type: array description: >- Tools the model may call. OpenAI function tool shapes are accepted. items: type: object tool_choice: description: >- Controls which tool the model calls. Pass none, auto, required, or a function shape. parallel_tool_calls: type: boolean description: Whether the model may call several tools in one turn. reasoning: type: object description: Reasoning controls, such as the effort level. store: type: boolean default: false description: >- Requests are not stored server-side; the field is accepted for compatibility. stream: type: boolean default: false description: >- If true, the response is streamed as server-sent events, matching the OpenAI Responses API: response.created, response.output_text.delta, and response.completed. stream_options: type: object description: Streaming options, such as include_usage. max_output_tokens: type: integer description: Maximum number of tokens to generate in the response. text: type: object description: Text output settings, such as a response format. metadata: type: object description: Metadata to attach to the response, in the OpenAI shape. previous_response_id: type: string description: >- The id of a previous response, for clients that chain responses client-side. include: type: array items: type: string description: >- Extra fields to include in the response, such as reasoning encodings. service_tier: type: string description: Accepted for compatibility. The gateway serves one tier. prompt_cache_key: type: string maxLength: 256 description: >- Routing hint that groups requests into one conversation only when no session header or body session id is present. See [Session affinity](https://subconscious.dev/features/session-affinity). session_id: type: string minLength: 8 maxLength: 256 description: >- Body session id, also spelled sessionId and providerOptions.gateway.sessionId. Between 8 and 256 characters. See [Session affinity](https://subconscious.dev/features/session-affinity). providerOptions: type: object description: >- Vercel AI SDK provider options. Only `gateway.sessionId` is read, as a body session id. See [Session affinity](https://subconscious.dev/features/session-affinity). return_subconscious_cache_stats: type: boolean description: >- Subconscious extension. Returns the worker's subconscious cache stats on the response. chat_template_kwargs: type: object description: >- Subconscious extension. Chat-template kwargs, such as the context budget. responses: '200': description: Successful completion content: application/json: schema: type: object properties: id: type: string description: A unique identifier for the response. example: resp_abc123 object: type: string enum: - response description: The object type. created_at: type: integer description: Unix timestamp of when the response was created. example: 1716000000 model: type: string description: The model used for the response. example: subconscious/glm-5.3-marathon output: type: array description: >- The response output items, such as messages and function calls. items: type: object prompt_cache_key: type: string description: >- Present only when the request sent one. Echoes the caller's value. x-codeSamples: - lang: python label: Python (OpenAI SDK) source: |- from openai import OpenAI client = OpenAI( api_key="your-api-key", base_url="https://api.subconscious.dev/v1", ) response = client.responses.create( model="subconscious/glm-5.3-marathon", input="What is the capital of France?", ) print(response.output_text) - lang: bash label: cURL source: |- curl https://api.subconscious.dev/v1/responses \ -H "Authorization: Bearer YOUR_API_KEY" \ -H "Content-Type: application/json" \ -d '{ "model": "subconscious/glm-5.3-marathon", "input": "What is the capital of France?" }' components: securitySchemes: bearerAuth: type: http scheme: bearer description: API key from your Subconscious dashboard ```` This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.