> ## Documentation Index > Fetch the complete documentation index at: https://docs.aigmented.io/llms.txt > Use this file to discover all available pages before exploring further. # Ask a question (RAG) > Retrieves relevant knowledge chunks from the collection and uses an LLM to generate a grounded answer. Supports both synchronous JSON responses and streaming via SSE (`stream: true`). Token usage is tracked and counted against the team's quota. ## OpenAPI ````yaml https://staging.aigmented.io/api/v1/openapi.json post /api/v1/collections/{id}/ask openapi: 3.1.0 info: title: Aigmented API version: 1.0.0 description: >- REST API for Aigmented — manage knowledge collections, upload files, search knowledge cards, and ask questions with LLM-powered answers. contact: name: Aigmented Support url: https://aigmented.com servers: - url: https://aigmented.io description: Production security: - BearerAuth: [] tags: - name: Collections description: Create and manage knowledge collections - name: Files description: Upload and manage source documents - name: Search description: Semantic search over knowledge cards in a collection - name: Ask description: LLM-powered question answering over a collection paths: /api/v1/collections/{id}/ask: post: tags: - Ask summary: Ask a question (RAG) description: >- Retrieves relevant knowledge chunks from the collection and uses an LLM to generate a grounded answer. Supports both synchronous JSON responses and streaming via SSE (`stream: true`). Token usage is tracked and counted against the team's quota. operationId: askCollection parameters: - name: id in: path required: true schema: type: integer description: Collection ID requestBody: required: true content: application/json: schema: $ref: '#/components/schemas/AskRequest' example: question: What does the agreement say about liability? model: gpt-4o-mini mode: fast stream: false top_k: 10 current_only: true responses: '200': description: >- Answer generated successfully. When `stream=false`, returns a JSON body. When `stream=true`, returns an SSE stream (`Content-Type: text/event-stream`). content: application/json: schema: $ref: '#/components/schemas/AskResponse' text/event-stream: schema: type: string description: >- Newline-delimited SSE events. Each event is a JSON object with `type` and `data` fields. See the `AskStreamEvent` schema for details. '400': description: Missing or invalid parameters content: application/json: schema: $ref: '#/components/schemas/Error' '401': description: Unauthorized content: application/json: schema: $ref: '#/components/schemas/Error' '404': description: Collection not found content: application/json: schema: $ref: '#/components/schemas/Error' '429': description: Team token limit exceeded content: application/json: schema: $ref: '#/components/schemas/TokenLimitError' '500': description: Internal server error content: application/json: schema: $ref: '#/components/schemas/Error' components: schemas: AskRequest: type: object required: - question properties: question: type: string description: The question to answer example: What does the document say about data retention? model: type: string default: gpt-4o-mini description: LLM model identifier to use for answering example: gpt-4o mode: type: string enum: - fast - full default: fast description: >- Answer mode. `fast` uses fewer retrieved chunks for a quicker response; `full` retrieves more context for a thorough answer. stream: type: boolean default: false description: >- If true, the response is streamed as Server-Sent Events (SSE). Each event has a `type` field: `delta` (text chunk), `done` (final metadata), or `error`. top_k: type: integer default: 10 minimum: 1 maximum: 50 description: Number of knowledge chunks to retrieve before generating the answer current_only: type: boolean default: false description: Restrict retrieval to the most current document versions only chat_history: type: array items: $ref: '#/components/schemas/ChatMessage' description: Optional prior conversation turns for multi-turn context example: - role: user content: Summarise the document. - role: assistant content: The document covers... filters: type: object description: Optional metadata filters to scope retrieval additionalProperties: true AskResponse: type: object properties: answer: type: string description: The LLM-generated answer sources: type: array items: $ref: '#/components/schemas/AskSource' description: Knowledge chunks used to generate the answer model: type: string description: Model used to generate the answer tokens_used: type: object description: Token usage breakdown properties: llm_prompt: type: integer llm_completion: type: integer embedding: type: integer model_id: type: string Error: type: object properties: error: type: string description: Human-readable error message required: - error TokenLimitError: type: object properties: error: type: string example: Token limit exceeded remaining: type: integer description: Remaining included tokens reason: type: string description: Reason for the limit required: - error ChatMessage: type: object required: - role - content properties: role: type: string enum: - user - assistant description: Message author role content: type: string description: Message text content AskSource: type: object properties: document_id: type: string file_name: type: string page: type: integer chunk_index: type: integer score: type: number format: float content_preview: type: string description: First ~200 characters of the source chunk securitySchemes: BearerAuth: type: http scheme: bearer bearerFormat: API key (sk-...) description: >- Pass your API key as a Bearer token. Example: `Authorization: Bearer sk-xxxxxxxxxxxx` ````