PerplexityapiBatch API for the Perplexity Agent API
Developers can now send up to 50,000 Agent API requests in one file and collect the results when the batch finishes, for developers on the Perplexity API.
Details
- You upload a file of up to 50,000 requests (200 MB max), create a batch, and download a results file when it finishes
- Each request runs the full Agent API loop, so built-in web search and page fetching work inside a batch
- Aims to finish within 24 hours. Requests not started within 48 hours expire and aren't billed
- Batch requests don't count toward the request rate limit. There's no separate batch price: OpenAI models run on their cheaper flex tier where supported, and other models are billed at the standard rate
- Every request in a file must use the same model. Custom function and connector tools aren't allowed, and it's not available to zero-data-retention organizations
Best for
Large research, enrichment, or classification jobs that don't need answers right away
When to use it
You have thousands of rows to run, like enriching a list of companies with web research, and don't want to manage rate limits request by request
Who benefits
Developers and data teams running bulk research jobs on Perplexity
Join the newsletter to get ahead with AI.
I share strategies, new modules, free guides, and tools to get you better at using AI in your work.
Leading a team? See team programs →