> ## Documentation Index
> Fetch the complete documentation index at: https://docs.requesty.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# OpenAPI Spec

> Download the Requesty Inference API OpenAPI specification

Download the OpenAPI 3.0 specification for the Requesty Inference API. Use it with your favorite API client, code generator, or testing tool.

<Card title="Inference API Spec" icon="download" href="/api-reference/requesty_inference-openapi.json">
  OpenAPI spec for inference endpoints (chat, embeddings, images, audio, models).
</Card>

## What's included

The Inference API spec covers all endpoints on `router.requesty.ai`:

| Endpoint                   | Method | Description                       |
| -------------------------- | ------ | --------------------------------- |
| `/v1/chat/completions`     | POST   | Create chat completion            |
| `/v1/messages`             | POST   | Create message (Anthropic format) |
| `/v1/responses`            | POST   | Create response (Responses API)   |
| `/v1/embeddings`           | POST   | Create embedding                  |
| `/v1/audio/speech`         | POST   | Create speech (TTS)               |
| `/v1/audio/transcriptions` | POST   | Create transcription (STT)        |
| `/v1/images/generations`   | POST   | Create image                      |
| `/v1/images/edits`         | POST   | Edit image                        |
| `/v1/models`               | GET    | List models                       |

## Web Search in the spec

The spec documents web search support across all three chat endpoints:

* **Chat Completions** (`Tool` schema): `"type": "web_search"` alongside `"function"`
* **Responses API** (`ResponsesTool` schema): `"type": "web_search"` alongside `"function"`
* **Messages API** (`AnthropicTool` schema): `"type": "web_search_20250305"` with `name` and `max_uses`

See the [Web Search guide](/features/web-search) for full usage examples and response formats.

## Usage examples

Import the spec into popular tools:

* **Postman**: Import > Link > paste the spec URL
* **Insomnia**: Application > Import > From URL
* **OpenAPI Generator**: `openapi-generator-cli generate -i <spec-url> -g <language>`
* **Swagger UI**: Paste the spec URL into the explore bar