Is the DeepSeek API reachable right now?
A live probe of api.deepseek.com from NetOkay’s edge, a test from your own browser, and why DeepSeek’s reasoning models are the most common way to discover an idle timeout on your own path.
In short: The card below shows the latest probe of api.deepseek.com from
eight Cloudflare regions: one GET for /models with a placeholder token and
expects the API’s own 401. Then test from this browser: if NetOkay reaches the host and
your reasoning requests die after half a minute, the likely cause is a proxy closing
silent connections during the model’s think phase.
Last updated
Live probe · api.deepseek.com
Where is api.deepseek.com reachable from?
Loading the latest observation of api.deepseek.com… Test from this browser →
From this browser
—Not run. One request from your browser, sent only when you ask.
What this proves, and what it does not
Loading…
DeepSeek specifics
- API host
-
api.deepseek.com, HTTPS on port 443, OpenAI‑compatible. The/v1prefix is optional;/modelsand/v1/modelsreach the same endpoint. - Streaming
-
OpenAI‑style server-sent events. For
deepseek-reasoner, chunks carryreasoning_contentbefore the finalcontent, and the stream ends withdata: [DONE]. Clients that ignorereasoning_contentsee a long apparent silence. - Think phases
- Reasoning can run for a minute or more. If your client does not consume reasoning chunks, or a middlebox counts them as idle because it buffers them, the connection is cut at the middlebox’s idle limit. The homepage idle stream check reproduces this without a key.
- Official status
- DeepSeek publishes service status at status.deepseek.com.
- Client defaults
-
DeepSeek’s docs point you at the OpenAI SDK with
base_url="https://api.deepseek.com", so you inherit its 10-minute timeout and 2 retries. A reasoning request cut at 30 seconds is therefore never the SDK; look at the path between you and api.deepseek.com.
What DeepSeek status codes mean
| Code | Sent by | Meaning |
|---|---|---|
| 401 | API application | Invalid key. The API application produced this answer. |
| 402 | API application | Insufficient balance, which is DeepSeek‑specific: top up before retrying. |
| 429 | API application | Rate limit reached. |
| 500 / 503 | API application | Server-side error or overloaded. A 400 means an invalid request body. |
| Every row is an HTTP answer: the host was reached and the request was refused or served. None of them means api.deepseek.com is unreachable. | ||
Reachable here but failing for you?
-
Run the 30‑second idle stream from the network where your app fails. A think phase with
unconsumed
reasoning_contentis silence to a proxy; if the idle stream is cut at 30 seconds, so is the reasoning request. Run the idle stream → -
Make your client consume
reasoning_contentchunks as progress, then read the idle result: closed during silence is an idle limit, start and done together is buffering. Fixes in the streaming guide → - Compare environments if this browser passes but a container, server or agent fails, then run the CLI from the failing environment against api.deepseek.com. Compare environments →
DeepSeek API status FAQ
My deepseek-reasoner request closes after about 30 seconds with no output. Is DeepSeek down?
Probably not. That timing matches common idle limits on proxies, gateways and some cloud load balancers. Run the homepage idle stream check from the same network; if it fails at a similar time, the limit is on your path.
What does 402 mean?
Insufficient balance on your DeepSeek account. The host is reachable and answered; top up or check billing.
Is the probe result specific to my region?
No. It comes from whichever Cloudflare location served your page request, refreshed at most once a minute. Your own route to api.deepseek.com can differ; that is why the browser checks exist.