DeepSeek API adds web search, file input and voice tools
The DeepSeek App's search, file, link, safety and voice features now work on deepseek-flash.
DeepSeek released its Full-Suite App API on November 13, 2026. The features behind the DeepSeek App (search, file input, link reading, safety checks and voice) now work through the DeepSeek API, on the deepseek-flash model. DeepSeek already sold Search and Fetch as standalone APIs (earlier coverage).
What changed
deepseek-flash(DeepSeek-V4.1-Flash-1017) now supports web search, deep thinking search, file input, URL reading and safety checks across the Messages, Responses and Chat Completions APIs. DeepSeek runs the tool calls and context assembly server-side.- Turn on search with the
deepseek_web_searchtool and link reading with thedeepseek_read_urltool. With thinking off, the model searches the web or images and answers with citations. With thinking on, it searches and reads pages over multiple reasoning steps. - Public-facing apps must enable safety checks (the
safety_checkparameter). DeepSeek says developers remain responsible for content moderation, model registration and AI-generated content labeling. - Search, Fetch, OCR, ASR (speech to text) and TTS (text to speech) are also available as standalone APIs.
- DeepSeek published an open-source chat frontend demo (
deepseek-ai/chat-demo) that can import DeepSeek chat history.
Tool prices
From DeepSeek's pricing page on November 13, 2026. Tool fees are charged on top of token charges. DeepSeek lists the speech discounts as limited-time and gives no end date.
| Tool | Unit | List price | Current price |
|---|---|---|---|
| Search (text, image) | 1,000 searches | $0.75 | $0.75 |
| Web fetch | 1,000 fetches | $0.75 | $0.75 |
| File parsing and OCR | 1,000 pages | $0.225 | $0.225 |
| Text to speech (TTS) | 10,000 characters | $0.45 | $0.225 |
| Speech to text (ASR) | 1 hour of audio | $0.116 | $0.058 |
DeepSeek's own token prices for deepseek-flash are $0.15 input, $0.003 cached input and $0.60 output per million tokens off-peak, and double that at peak (01:00 to 04:00 and 06:00 to 10:00 UTC, Monday to Friday), per the same page.
DeepSeek V4.1 Flash 1017 on RouterPlus
DeepSeek documents the built-in tools for its own API at api.deepseek.com. For model tokens only, the RouterPlus catalog lists DeepSeek V4.1 Flash 1017 as deepseek/deepseek-v4.1-flash-1017 on 7 hosts. Per million tokens, from the catalog on November 13, 2026:
| Provider | Input | Cached input | Output |
|---|---|---|---|
| Sail Research | $0.09 | $0.006 | $0.30 |
| Wafer | $0.13 | $0.008 | $0.50 |
| Doubleword | $0.15 | $0.003 | $0.60 |
| RunInfra | $0.15 | $0.01 | $0.60 |
| Particle AI | $0.2125 | $0.0255 | $0.85 |
| Munito AI | $0.27 | $0.0054 | $1.00 |
| Nebius | $0.30 | $0.006 | $1.20 |
What to do
To try the tools on DeepSeek's API, add them to a deepseek-flash request. From DeepSeek's chatbot guide (Messages API):
curl https://api.deepseek.com/anthropic/v1/messages \
-H "Content-Type: application/json" \
-H "x-api-key: ${DEEPSEEK_API_KEY}" \
-d '{
"model": "deepseek-flash",
"thinking": {"type": "enabled"},
"tools": [
{"type": "deepseek_web_search", "name": "deepseek_web_search"},
{"type": "deepseek_read_url", "name": "deepseek_read_url"}
],
"safety_check": {"level": "high"},
"messages": [{"role": "user", "content": "What is new in DeepSeek V4.1?"}]
}'