Search API

LLMGateway exposes a /v1/search endpoint compatible with the Perplexity Search API. It returns ranked web results with extracted page snippets, not a model-written answer. Use it for your own retrieval, grounding, or agent tools. If you want a model to search and answer in one call, use native web search instead.

Existing Perplexity Search code works after you swap the base URL and API key.

For the full request and response schema, see the API reference.

Endpoint

POST https://api.llmgateway.io/v1/search

cURL

curl -X POST "https://api.llmgateway.io/v1/search" \
  -H "Authorization: Bearer $LLM_GATEWAY_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "query": "latest developments in open-source LLMs",
    "max_results": 3,
    "search_recency_filter": "week"
  }'

{
	"id": "9ed15dce-a498-40b2-8bc9-7f13f35901cd",
	"model": "perplexity/perplexity-search",
	"results": [
		{
			"title": "…",
			"url": "https://…",
			"snippet": "…",
			"date": "2026-09-25",
			"last_updated": "2026-09-27"
		}
	],
	"server_time": null
}

Models

model is optional. When it is omitted, search_type picks the model:

Sending both a model and a search_type that disagree returns a 400. search_type: "people" is not supported yet. Current pricing is on each model's page on the models page.

Request fields

Every field except model is forwarded to Perplexity unchanged.