> ## Documentation Index > Fetch the complete documentation index at: https://docs.corti.ai/llms.txt > Use this file to discover all available pages before exploring further. # Create chat completion > Creates a model response for the given conversation. This is the OpenAI Chat Completions API. Some models return chain-of-thought reasoning in the `reasoning` field of the response message. The `-instant` model variants do not produce reasoning on this endpoint. ## OpenAPI ````yaml /api-reference/corti-models/corti-models-openapi.yml post /chat/completions openapi: 3.0.3 info: title: Corti S1 API description: > OpenAI-compatible API for Corti's LLM deployment. The API exposes two endpoints: * **`/v1/chat/completions`** - OpenAI Chat Completions API. * **`/v1/responses`** - OpenAI Responses API. Supports `instructions` as an alternative to system messages. Both endpoints support text generation, streaming (SSE), tool/function calling, and JSON mode. ## Available Models | Model | Description | |------|-------------| | `corti-s1` | Full-capability model | | `corti-s1-instant` | Fast variant of corti-s1 | | `corti-s1-mini` | Smaller model | | `corti-s1-mini-instant` | Fast variant of corti-s1-mini | `-instant` variants are optimized for speed and lower token usage. ## Key Characteristics * **No embeddings, vision, or file support**: Only text input/output. * **`reasoning` field**: Some models return chain-of-thought reasoning in a `reasoning` field (chat completions) or as structured output items with `type: "reasoning"` (responses API). The `-instant` variants do not produce reasoning on chat completions. version: 1.0.0 contact: name: Corti AI url: https://corti.app servers: - url: https://ai.eu.corti.app/v1 description: Production (EU) security: - bearerAuth: [] paths: /chat/completions: post: tags: - Chat summary: Create chat completion description: > Creates a model response for the given conversation. This is the OpenAI Chat Completions API. Some models return chain-of-thought reasoning in the `reasoning` field of the response message. The `-instant` model variants do not produce reasoning on this endpoint. operationId: createChatCompletion requestBody: required: true content: application/json: schema: $ref: '#/components/schemas/ChatCompletionRequest' examples: Basic: summary: Basic chat completion value: model: corti-s1 messages: - role: user content: Count from 1 to 5 max_tokens: 500 Streaming: summary: Streaming response value: model: corti-s1 messages: - role: user content: Count from 1 to 5 max_tokens: 500 stream: true JSON Mode: summary: Structured JSON output value: model: corti-s1 messages: - role: user content: >- Return a JSON object with fields name and age for John who is 30. response_format: type: json_object max_tokens: 500 Tool Calling: summary: Function/tool calling value: model: corti-s1 messages: - role: user content: What is the weather in Copenhagen today? tools: - type: function function: name: get_weather description: Get weather information for a city parameters: type: object properties: city: type: string required: - city tool_choice: auto max_tokens: 500 Multi-turn: summary: Multi-turn conversation value: model: corti-s1 messages: - role: user content: Count from 1 to 5 - role: assistant content: 1, 2, 3, 4, 5 - role: user content: Now count from 10 to 15 max_tokens: 500 responses: '200': description: > Successful response. For non-streaming requests, returns a complete chat completion. For streaming requests, returns Server-Sent Events (SSE) with incremental chunks. content: application/json: schema: $ref: '#/components/schemas/ChatCompletionResponse' text/event-stream: schema: $ref: '#/components/schemas/ChatCompletionStreamResponse' '401': description: Authentication failed content: application/json: schema: $ref: '#/components/schemas/Error' '429': description: Rate limit exceeded content: application/json: schema: $ref: '#/components/schemas/Error' components: schemas: ChatCompletionRequest: type: object required: - model - messages properties: model: type: string enum: - corti-s1 - corti-s1-instant - corti-s1-mini - corti-s1-mini-instant default: corti-s1 messages: type: array minItems: 1 description: > Conversation messages. Supported roles: `user`, `assistant`, `tool`, and `system`. items: $ref: '#/components/schemas/Message' stream: type: boolean default: false description: > If true, returns SSE with `data: {chunk}` lines. Terminates with `data: [DONE]`. temperature: type: number format: float minimum: 0 maximum: 1 default: 1 top_p: type: number format: float minimum: 0.01 maximum: 1 default: 0.95 max_tokens: type: integer minimum: 1 description: Maximum tokens to generate. tools: type: array description: Function tools the model may call. Max 128. items: $ref: '#/components/schemas/FunctionTool' tool_choice: type: string enum: - auto default: auto response_format: type: object properties: type: type: string enum: - json_object required: - type logprobs: type: boolean default: false top_logprobs: type: integer minimum: 0 maximum: 20 ChatCompletionResponse: type: object properties: id: type: string object: type: string enum: - chat.completion created: type: integer model: type: string description: The model name used for this response. choices: type: array items: type: object properties: index: type: integer message: $ref: '#/components/schemas/ChatResponseMessage' logprobs: type: object nullable: true finish_reason: type: string enum: - stop - length - tool_calls stop_reason: type: integer nullable: true description: Internal stop reason code from the inference engine. token_ids: type: array nullable: true items: type: integer routed_experts: type: array nullable: true items: {} description: Only present on corti-s1 and corti-s1-instant. system_fingerprint: type: string nullable: true usage: type: object properties: prompt_tokens: type: integer completion_tokens: type: integer total_tokens: type: integer prompt_tokens_details: type: object nullable: true service_tier: type: string nullable: true prompt_logprobs: type: object nullable: true prompt_token_ids: type: array nullable: true items: type: integer prompt_text: type: string nullable: true kv_transfer_params: type: object nullable: true ChatCompletionStreamResponse: type: object description: 'SSE chunk. Stream terminates with `data: [DONE]`.' properties: id: type: string object: type: string enum: - chat.completion.chunk created: type: integer model: type: string choices: type: array items: type: object properties: index: type: integer delta: type: object properties: role: type: string content: type: string nullable: true reasoning: type: string nullable: true finish_reason: type: string nullable: true enum: - stop - length - tool_calls Error: type: object required: - error properties: error: type: string description: Error message or object with message. detail: type: string Message: oneOf: - title: User Message type: object required: - role - content properties: role: type: string enum: - user content: type: string - title: System Message type: object required: - role - content properties: role: type: string enum: - system content: type: string - title: Assistant Message type: object required: - role properties: role: type: string enum: - assistant content: type: string nullable: true reasoning: type: string nullable: true description: > Chain-of-thought reasoning, if produced by the model. The `-instant` variants do not produce this field. tool_calls: type: array items: $ref: '#/components/schemas/ToolCall' - title: Tool Message type: object required: - role - content - tool_call_id properties: role: type: string enum: - tool content: type: string tool_call_id: type: string FunctionTool: type: object required: - type - function properties: type: type: string enum: - function function: type: object required: - name - description - parameters properties: name: type: string pattern: ^[a-zA-Z0-9_-]+$ maxLength: 64 description: type: string parameters: type: object ChatResponseMessage: type: object properties: role: type: string enum: - assistant content: type: string nullable: true reasoning: type: string nullable: true description: > Chain-of-thought reasoning, if produced by the model. The `-instant` variants return `null` for this field. refusal: type: string nullable: true annotations: type: array nullable: true items: {} audio: type: object nullable: true function_call: type: object nullable: true tool_calls: type: array items: $ref: '#/components/schemas/ToolCall' ToolCall: type: object properties: id: type: string type: type: string enum: - function function: type: object properties: name: type: string arguments: type: string description: JSON-encoded arguments. securitySchemes: bearerAuth: type: http scheme: bearer description: > Use a Corti-issued bearer token. The token is a Base64-encoded client credential string in the format `base:client_credentials::`. ````