Exa Deep Research FAQs

Question 1

Is Exa Research more expensive than standard search?

Accepted Answer

Yes, research is a premium feature that uses reasoning tokens and page reads. It is best used for complex tasks where standard search or basic summaries are insufficient.

Question 2

Can I get output in a specific format?

Accepted Answer

Yes, you can provide a JSON schema using the --output-schema parameter to ensure the research results are structured exactly how you need them for your workflow.

Question 3

How does Exa Research differ from standard web search?

Accepted Answer

Unlike standard search which provides immediate links, research is an asynchronous, multi-step process that uses AI reasoning to synthesize information and can follow a specific output schema.

Question 4

Which research models are available?

Accepted Answer

The skill supports exa-research (default) for a balance of speed and quality, exa-research-pro for comprehensive results, and exa-research-fast for simpler queries.

Question 5

How do I check the status of a long-running research task?

Accepted Answer

You can use the research-get command with your unique research ID to check progress or retrieve the final results once the task status is marked as completed.

Exa Deep Research

Key Features

Use Cases

Exa Deep Research

Key Features

Use Cases