How to Set Up DeepSeek on Janitor AI?

[tessalockwood]

I have been attempting to come out of the normal models and I heard that great things are done by employing OpenRouter or even the official API to incorporate improved LLMs. Can anyone provide a step-by-step guide on how to set up deepseek on janitor ai so I can get more coherent and uncensored responses? I am particularly confused with what proxy URL, as well as what the recommended proxy Generation Settings such as temperature, and token limits would be with this particular model.

 

[Dorian]

I've managed to set up DeepSeek on Janitor AI. First, download the Janitor AI model and DeepSeek module. Then, follow the official DeepSeek documentation to configure the settings. Next, update the Janitor AI configuration file with the DeepSeek settings. Finally, restart the Janitor AI server and DeepSeek should be active. If you encounter issues, check the Janitor AI logs for error messages.

 

[Percival]

In order to install DeepSeek on Janitor AI, you will need to create an account on DeepSeek and create an API key. Then hit the Janitor AI settings, then in API or Custom API, paste the key and fill the base URL. Choose such a model as deepseek-chat, save, and allocate it to a character and start chatting.

 

[Geralt_of_Rivia]

To set up DeepSeek on Janitor AI, first create an account on DeepSeek and generate your API key from the dashboard, then log in to Janitor AI, go to settings or API configuration, choose the option to use your own API, select a compatible OpenAI-format model (if DeepSeek supports it), paste your API key, set the correct base URL provided by DeepSeek, save the settings, and test the connection to make sure it’s working properly.

 

[Rolf]

To set up DeepSeek on Janitor AI easily, first create an account with a model provider like DeepSeek or OpenRouter and generate your API key. Then open Janitor AI, go to API/Proxy Settings, choose Custom or Proxy, and paste your DeepSeek API key along with the proxy URL (e.g., https://openrouter.ai/api/v1/chat/completions). Enter the correct model name (like deepseek-chat or another DeepSeek model), save, and refresh. Once saved, your chats in Janitor AI will use DeepSeek for responses.