Overview

The Research API helps you ask complex questions and receive structured answers backed by citations. Under the hood, Exa agents:

1. Break your instructions into sub-tasks.
2. Gather relevant sources from the public web.
3. Extract key facts and reason over them with an LLM.
4. Return JSON that exactly matches the schema you provided.

Because tasks are asynchronous, you submit a request and immediately receive a task id. Poll the task until it is complete or failed, or list all tasks to monitor progress in bulk.

​

Core Endpoints

​

Typical Flow

1. Create – Submit natural-language instructions and an output schema.
2. Poll – Periodically call GET /research/tasks/{id} until the status is completed.
3. Consume – Use the structured output returned by the task in your application.

You can also call GET /research/tasks to retrieve a paginated list of tasks and their statuses.