Quickstart
Start querying the web with Parallel Task API v1
1. Set up Prerequisites
Generate your API key on Platform. Then, set up with the Python SDK or with cURL:
2. Execute your First Task Run
Make your first API request with one of these examples:
Sample Response
Immediately after a Task Run is created, the Task Run object, including the status of the Task Run, is returned. On completion, the Task Run Result object is returned.
Basis, including citations, reasoning - and for advanced Processors excerpts and confidence level - is returned with every Task Run Result.
3. From Simple to Complex Tasks
The Task API supports a progression of task complexity:
Simple Query -> Simple Answer
Simply ask “What was the founding date of the United Nations?” and receive an answer based on web research in plain text. This straightforward approach is illustrated above.
Simple Query -> Structured Output
Define a structured output format with founding date, employee count and other desired details for a company, passing in the company name as input.
Structured Input -> Structured Output
Define structured input and output schemas providing founding date, employee count and other desired details for a company name and company website.
Sample Structured Task Result
4. Run Multiple Tasks
For many use cases, knowing how to scale beyond one Task Run is crucial. To execute multiple Task Runs concurrently, we recommend the AsyncParallel Python client: