Back to the feedVercel Gateway Deals
Ends Jan 31

Parallel search normally $5 per 1,000 requests

Parallel web search is free on Vercel AI Gateway

Parallel web search in fast mode is free on AI Gateway through January 31, 2027 (normally $5 per 1,000 requests)

Parallel web search on Vercel AI Gateway costs nothing in fast mode, instead of the usual $5 per 1,000 requests. It works with any AI Gateway model and needs no separate Parallel API key. The free period runs through January 31, 2027.

Eligible plans
Vercel AI Gateway pay-as-you-go
Eligibility
Anyone with a Vercel account and AI Gateway; no separate Parallel API key needed; fast mode only
Benefit duration
Free through January 31, 2027
Renewal terms
After January 31, 2027, fast mode returns to $5 per 1,000 requests; other modes are always billed at standard pricing
Claim by
2027-01-31

The briefing

Vercel has brought Parallel web search to AI Gateway, and searches in fast mode are completely free through January 31, 2027. If you are adding web access to an agent or chat app, this is a good window to search as much as you need.

How much you save

Parallel search normally costs $5 per 1,000 requests, with up to 10 results per request, plus $1 per 1,000 additional results. On AI Gateway, fast-mode searches are billed at $0, including results beyond 10. Only fast mode is free; other modes are billed at standard pricing, and model usage is billed as usual.

After January 31, 2027, fast mode returns to the regular price. Other Parallel prices stay the same.

What you can use

Parallel finds relevant web pages and returns links with short excerpts, and you can filter results by website or publication date. It works with any AI Gateway model, with no separate Parallel API key. Free searches have Zero Data Retention enabled by default, so Parallel does not retain the data or use it for training.

If you build agents with Vercel's eve, the built-in web_search tool now uses Parallel fast mode by default and is free when the model is served through AI Gateway.

How to use

  1. Sign in to Vercel and create an API key in AI Gateway.
  2. With the AI SDK, add gateway.tools.parallelSearch({ mode: 'fast' }) to your tools. With Chat Completions, set mode to fast in the config of the vercel:parallel_search tool.
  3. Send requests as usual, and the model calls the search tool when it needs current information.

See the Vercel announcement and the AI Gateway web search docs.

Is this still live?

Community signalAwaiting first signal