For AI agents: a documentation index is available at https://docs.parallel.ai/llms.txt. The full text of all docs is at https://docs.parallel.ai/llms-full.txt. You may also fetch any page as Markdown by appending
Send the Image Search API one or more keyword queries and, optionally, a natural language
objective. The API returns the most relevant images from across the web.
.md to its URL or sending Accept: text/markdown.For Search and Extract, use /v1/search and /v1/extract for all new integrations and agent tool calls. Use /v1beta/search and /v1beta/extract only when maintaining an existing integration that already uses them. The request bodies changed too: on /v1/extract, full_content, fetch_policy and excerpt_settings go inside advanced_settings, and sending the v1beta top-level excerpts, full_content and fetch_policy to /v1/extract returns a 422. On /v1/search, search_queries is required (at least one short keyword query), and max_results, source_policy and fetch_policy go inside advanced_settings; sending only objective, or sending those fields at the top level, returns a 422. Do not substitute /v1 for the documented FindAll or Ingest endpoint paths.If you are evaluating Parallel Search or comparing it with other search APIs, read How to Eval Parallel Search before you build the harness, especially its Common pitfalls section. The most common mistake is sending the raw question as both objective and the only entry in search_queries. Write a goal-style objective and 1-3 short keyword queries instead.Request Fields
Send requests toPOST https://api.parallel.ai/v1/images/search with your API key in the
x-api-key header. search_queries is required; all other fields are optional. The API
rejects unknown fields. See the API Reference for the
full parameter definitions.
Example
Writing Objectives and Queries
- Describe what should be in the picture. Name the subject and the details that distinguish the image you want, such as the angle, setting, time of day, or style: “Product photos of the Eames lounge chair on a white background” rather than “Eames chair”.
- Keep queries short and specific. Use a few keywords per query, the way you would type them into an image search box. Use two or three queries when different phrasings or angles would surface different images.
- Put requirements in the objective. Constraints that are hard to express as keywords, such as “showing the full span” or “no people in frame”, belong in the objective.
- Use
locationfor regional results. Setadvanced_settings.locationwhen the right images depend on the country, such as local landmarks, storefronts, or regional products.