DeepSeek documents Search and Fetch APIs at $0.75 per 1,000

DeepSeek is making live search and page retrieval a cheap building block for agent stacks, but teams will still want to avoid redundant Fetch calls and plan around separate Search and Fetch rate limits.

DeepSeek has published docs for two web-retrieval endpoints for agents: POST /v1/tools/search and POST /v1/tools/fetch, both exposed as raw JSON APIs via the DeepSeek guide.

Search returns structured web results, while Fetch pulls one URL and returns the page body in Markdown. If you set need_content on Search, DeepSeek can attach page content directly to results, which means some flows may not need a separate Fetch call.

Both APIs are priced at $0.75 per 1,000 successful requests. Search is billable only when it returns HTTP 200 with a non-empty results array; Fetch is billable only when page content is returned. DeepSeek also lists separate account-level caps of 5 QPS for Search and 5 QPS for Fetch.

The docs also note that server-side tool calls through DeepSeek Harness, Claude Code, or Codex are free for now, with per-request billing starting at 00:00 Beijing time on November 1, 2026.

Build an AI agent

Sources

DeepSeek guide

When models and APIs shift keep the workflow running

Connect the models and apps from stories like this in one scenario. If a vendor changes routing, pricing, or availability, you update the workflow — you don't start over.

Start Free

Free forever plan. No credit card required.