Skip to content

DeepSeek opens a free search API for Claude Code and Codex agents

The Search and Fetch endpoints, released alongside DeepSeek-V4.1-Flash-1017, cost $0.75 per 1,000 calls but run free through DeepSeek Harness, Claude Code and Codex until 31 October.

Source: DeepSeek

DeepSeek released DeepSeek-V4.1-Flash-1017 on 17 October, an update to its Flash model, and used the same note to open its Search and Fetch APIs to everyone. The company describes the model update as “a Pareto improvement in performance and cost” over the previous Flash release, with faster task completion and shorter outputs. Weights are open-sourced on Hugging Face and ModelScope, and DeepSeek says the model is optimised for its own Harness (v0.2.1-rc.1 or later), plus Tencent’s CodeBuddy and WorkBuddy.

The bigger news for agent builders is the Search and Fetch endpoints themselves. They run on DeepSeek’s own search engine, which the company says already serves hundreds of millions of requests daily. Pricing is $0.75 per 1,000 calls per endpoint, but calls made through server-side tools in DeepSeek Harness, Claude Code and Codex are free for DeepSeek API users until billing starts on 31 October 2026 at 16:00 UTC.

Why it matters: if you’ve wired DeepSeek into Claude Code or Codex as a tool provider, you get a working web search and fetch tool at no cost for two weeks. After 31 October it becomes a metered call, so anything built against it now needs a line in your cost model before then.

The caveat: DeepSeek’s release note doesn’t give token pricing, licence terms or context window for V4.1-Flash-1017 itself in text, its benchmark comparisons are shown only as images, so none of those numbers appear here.

Share

Related reading

ML for Web Devs

Tokens, context windows, and why your bill is what it is

The unit of billing for language models is not the request or the word. Once you understand tokens and how a context window fills up, the invoice stops being a surprise.

ML for Web Devs

RAG is just search with extra steps

Retrieval-augmented generation gets written up like an architecture. It is a search query, a string concatenation and one API call, and the search query is the part that decides whether it works.

ML for Web Devs

A prompt is a spec, so write it like one

Prompt engineering has very little to do with magic words. It is the same skill as writing a clear ticket for a capable contractor who will not ask you any questions.

More from The Wire

All briefs

GitHub Copilot retires six AI models today

Gemini 3.7 Flash, GPT-5.5, GPT-5.4, GPT-5.4 mini, GPT-5 mini and Grok 4.5 stop working across Copilot Chat, inline edits, agent mode and completions from today.

via GitHub