|  | Introducing the Parallel Images API |
|
| Image Search designed for AI | Today we're launching Parallel Image Search, the first image search API built for agents. | Agents now build presentations, documents, and apps, and those often need the right image, not just the right words. Image Search gives them reliable, affordable access to images from across the web. | Like the Parallel Search API, it takes two inputs: | Keyword queries that define what to look for An objective that describes, in plain English, what a good result looks like and what to leave out
| That's the difference between getting the Ramp (company) logo and getting a photo of a loading ramp. Each request returns up to 20 images, and every result includes a direct image link, the page it came from, a title, and its dimensions, so agents can cite their sources and users can check them. | | What teams are building with the Images API | Image Search has been in early access since October. Teams are using it to: | Add the right headshots and company logos to research briefs, memos, and presentations Bulk gather headshots and logos for CRMs and company databases Find current product and merchant photos for shopping and delivery apps Pull visual references so app-building agents can pick colors and style before they build Source real-world reference photos for image and video generation
| Image Search is available today via the Parallel API and MCP. Read the quickstart guide. | | |
|