> ## Documentation Index > Fetch the complete documentation index at: https://aisa.one/docs/llms.txt > Use this file to discover all available pages before exploring further. # OpenAI Web Search > Ask a question and get an answer that an OpenAI model wrote after searching the live web. Send a Responses request — an `input` string (or message array) — and the endpoint injects a fixed, server-pinned model and the `web_search` tool for you; the model, tool and per-request search cap are server-controlled to keep cost bounded. The reply is a standard OpenAI Responses object: `output[]` contains `web_search_call` items (each a search that ran) and a `message` item with the answer text and URL citations, and the billed search count equals the number of `web_search_call` items. Billing is pay-as-you-go at exact cost: `web_search_calls × $0.01` plus the model's own token cost (fresh input = `input_tokens − cached`), with no markup. Use this when you want a written, cited answer grounded in current web content from an OpenAI model — for the Anthropic-model equivalent see [`post_anthropic_websearch_search`](/api-reference/search/post_anthropic-websearch-search), and for raw ranked links with extracted page text use [`post_tavily_search`](/api-reference/search/post_tavily-search). ## OpenAPI ````yaml openapi/zh/openai-web-search.json POST /apis/v1/openai-websearch/search openapi: 3.1.0 info: description: '' title: OpenAI Web Search (GEO) version: '1' x-aisa-capabilities: idempotency: Idempotency-Key max_price: X-AISA-Max-Price-USD quote: header: X-AISA-Cost-Mode value: quote x-aisa-configured-paths: - /apis/v1/openai-websearch/search x-aisa-document: facts_hash: c1d97f67687b6da5772cfc632b194d7c656654ee369b6ee6e176059c8212e94b generator_version: '2' protocol_version: '1' schema_version: '1' composer_version: '13' response_pending: [] document_hash: sha256:af35d4bb94c9a4835cd81c03320b7e57e821a657621b57966da0090f13b450ae x-aisa-plans: builder: 1 display_plan: payg payg: 1 similarweb_payg: 1 team: 1 version: e70a0959cd1b93b84ae16dffd7a538273ddb9c1d46277b82823c37ff1b6bc0fb x-aisa-provider: openai-web-search x-aisa-catalogs: openai-web-search: title: OpenAI Web Search (GEO) description: '' x-aisa-document: facts_hash: c1d97f67687b6da5772cfc632b194d7c656654ee369b6ee6e176059c8212e94b generator_version: '2' protocol_version: '1' schema_version: '1' x-aisa-plans: builder: 1 display_plan: payg payg: 1 similarweb_payg: 1 team: 1 version: e70a0959cd1b93b84ae16dffd7a538273ddb9c1d46277b82823c37ff1b6bc0fb x-aisa-capabilities: idempotency: Idempotency-Key max_price: X-AISA-Max-Price-USD quote: header: X-AISA-Cost-Mode value: quote servers: - url: https://api.aisa.one security: - bearerAuth: [] paths: /apis/v1/openai-websearch/search: post: tags: - websearch summary: OpenAI Web Search description: >- Ask a question and get an answer that an OpenAI model wrote after searching the live web. Send a Responses request — an `input` string (or message array) — and the endpoint injects a fixed, server-pinned model and the `web_search` tool for you; the model, tool and per-request search cap are server-controlled to keep cost bounded. The reply is a standard OpenAI Responses object: `output[]` contains `web_search_call` items (each a search that ran) and a `message` item with the answer text and URL citations, and the billed search count equals the number of `web_search_call` items. Billing is pay-as-you-go at exact cost: `web_search_calls × $0.01` plus the model's own token cost (fresh input = `input_tokens − cached`), with no markup. Use this when you want a written, cited answer grounded in current web content from an OpenAI model — for the Anthropic-model equivalent see [`post_anthropic_websearch_search`](/api-reference/search/post_anthropic-websearch-search), and for raw ranked links with extracted page text use [`post_tavily_search`](/api-reference/search/post_tavily-search). operationId: post_openai_websearch_search requestBody: required: true content: application/json: schema: type: object required: - input properties: input: type: string description: >- Your question or instruction, same format as the OpenAI Responses API `input`. A message array is also accepted. example: >- What are the three biggest AI announcements this week? Give one sentence each with sources. max_output_tokens: type: integer description: >- Optional cap on the number of tokens generated in the answer. example: 1024 instructions: type: string description: >- Optional high-level instructions to steer the answer's tone or format. example: input: >- What are the three biggest AI announcements this week? Give one sentence each with sources. max_output_tokens: 1024 responses: '200': description: >- A standard OpenAI Responses object. `output[]` interleaves web_search_call items and the cited answer message; the number of web_search_call items is the billed search count. content: application/json: schema: type: object properties: id: type: string object: type: string example: response status: type: string example: completed output: type: array description: >- Ordered items: web_search_call (searches performed) and message (the cited answer). items: type: object usage: type: object properties: input_tokens: type: integer output_tokens: type: integer input_tokens_details: type: object properties: cached_tokens: type: integer default: content: application/json: schema: {} description: >- Error response; upstream passthrough responses may use provider formats components: securitySchemes: bearerAuth: scheme: bearer type: http ```` This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.