Apps
JanitorAI proxy setup
JanitorAI’s proxy setting takes any OpenAI-compatible API. It wants the full address, ending in /chat/completions: that one detail is behind most failed setups.
Checked 26 October 2026
The values
- Proxy URL
https://nokeep.ai/v1/chat/completions- API key
sk-nokeep-test to test, then your own sk-nokeep-… key- Model
deepseek-v4.1-flash-uncensored
Start with the free test key: it answers one fixed line, which proves the setup is right without spending anything. Then paste your own key from the console. deepseek-v4.1-flash-uncensored for most chats and pictures, or glm-5.3-flash-uncensored for text at under half the price.
Connect
- Open any chat and open its API settings, from the button at the top of the chat.
- Choose Proxy, and add a new configuration.
- Type
deepseek-v4.1-flash-uncensoredas the model. - Paste
https://nokeep.ai/v1/chat/completionsas the proxy URL, with/chat/completionson the end. - Paste
sk-nokeep-testas the API key. - Save, then refresh the page: JanitorAI picks up a new proxy only after a refresh.
- Send a message. A reply that starts “Connected to” means the setup works.
- Swap in your own key from the console, save and refresh again.
JanitorAI needs the full URL. https://nokeep.ai/v1 on its own answers 404 here.
Settings worth knowing
- JanitorAI calls the API from your browser. nokeep allows that, so no extra proxy is needed.
- DeepSeek answers at once in JanitorAI: it thinks only when a request asks, and JanitorAI doesn’t ask.
- Keep any custom prompt you like, or leave it empty. Nothing of ours is added to what JanitorAI sends.
- A bigger context setting sends more of the chat each time, which costs more per reply.
If it doesn’t connect
| The app shows | Fix |
|---|---|
| 401 | The key is wrong or revoked. Paste it again, with no spaces or quotes. |
| 404 | The proxy URL is missing /chat/completions. Use https://nokeep.ai/v1/chat/completions. |
| 402 | Your account is out of bought credits. Top up in the console, from $5. |
| 429 | Over 120 requests a minute, or the key’s daily cap. Raise the cap in the console, or wait. |
| 503 | The model is busy. Send again in a few seconds; nothing was charged. |
| The fixed test line | The setup works. Swap in your own key to get real replies. |
Every error is listed with its fix in Errors.
What an evening costs
At 3,000 tokens in and 250 out a reply, $5 is about 2,700+ replies on DeepSeek, or 6,500+ replies on GLM. A longer character card or a longer chat history costs more per reply, since the whole history is sent each time. Work out your own.
Questions
Is this the same as JanitorAI’s own model?
No. With a proxy, JanitorAI sends your chat to the model you set, here DeepSeek V4.1 Flash or GLM 5.3 Flash with their refusals removed, billed to your nokeep credits.
Can JanitorAI see my nokeep key?
The key is saved in your JanitorAI proxy settings and sent from your browser. Give JanitorAI its own key with a daily cap, so it can only spend so much.