Are you tired of JanitorLLM ruining you for everyone else? Or is your character forgetting important details during roleplay? Then it’s time for you to start using DeepSeek on JanitorAI.
JanitorAI’s in-house JanitorLLM (JLLM) is free to use and has no usage limits or filters, which is great. However, its context size (memory) is limited, and its responses become repetitive and boring fairly quickly.
Why Use DeepSeek On JanitorAI?
Other larger LLM models, like DeepSeek, are trained on a broader range of data. They have larger context sizes (memory), generate more creative responses, and follow instructions more effectively. You can enhance your roleplaying experience by using DeepSeek on JanitorAI.
What Is DeepSeek?
DeepSeek is a Chinese AI company that has created several open-source LLMs, including DeepSeek R1, DeepSeek V3.1, and DeepSeek V3.2, which is their latest model.
Also Read: DeepSeek V3.2’s Performance In AI Roleplay
The release of DeepSeek-R1 in early 2025 shook the AI world, as it was an open-source reasoning model developed at a lower cost than other LLMs. Its performance also matched top-tier models like OpenAI’s GPT-4 in reasoning and coding tasks.
What Are The Different DeepSeek Models?
DeepSeek R1 is designed for complex problem-solving and explains its chain of thought (CoT) before answering, which helps it generate better responses. In contrast, DeepSeek V3 is a chat model created for simple tasks and does not reason before replying.
On August 21, 2025, DeepSeek launched DeepSeek V3.1, a hybrid model that supports both thinking and non-thinking modes. On September 29, 2025, DeepSeek launched DeepSeek V3.2 with DSA, significantly improving long-context training and inference efficiency with minimal impact on output quality.
While you can still use DeepSeek’s older models from other providers, DeepSeek has replaced them with its new V3.2 model on its official API.
Use DeepSeek On JanitorAI Via The Official API [Paid]
You can use DeepSeek on JanitorAI directly through the official API. To get started, you need to deposit a minimum of $2 into your DeepSeek account. DeepSeek deducts funds from your account balance based on token usage (input and output).
Also Read: Understanding Tokens And Context Size
DeepSeek used to offer special discounted hours (off-peak pricing), during which prices dropped by up to 75 percent. They stopped offering discounted hours after September 5th, 2025. Even without the discounted hours, the official DeepSeek API remains the most affordable paid option if you value stability, speed, and quality.
Configure DeepSeek Official API As A Proxy On JanitorAI
- Register: Create an account with DeepSeek.
- Top Up: Add funds to your DeepSeek account with at least $2.
- API Key: Create an API key and save it in Notepad, as it is only shown to you once during creation.
Start a new chat or open an existing one with any character on JanitorAI, then click on the “using janitor” button on the menu bar at the top to access the API settings menu.
Click on the “Proxy” option in the API settings menu and then on “+ Add Configuration.” There are three key options for configuring DeepSeek on JanitorAI.
- Model:
deepseek-reasonerORdeepseek-chat - Proxy URL:
https://api.deepseek.com/v1/chat/completions - API Key: Paste the API key you created on DeepSeek’s site.
Make sure there are no unwanted spaces or typos in your input. It’s best to copy and paste these details instead of typing them manually. Save your settings and refresh your chat page. You should now be ready to use DeepSeek on JanitorAI.
Use DeepSeek on JanitorAI With OpenRouter [Free]
You can use DeepSeek on JanitorAI with OpenRouter, a proxy service that offers a unified interface for accessing different free and paid LLM models.
OpenRouter lets you send 50 free messages daily to their free DeepSeek models. You can add $10 to your OpenRouter account to increase the daily free message limit to 1,000.
Use OpenRouter To Set Up DeepSeek On JanitorAI
- Register: Sign up for an OpenRouter account.
- API Key: Create an API Key and save it in Notepad, as it’s only visible to you once during creation.
Start a new chat or open an existing one with any character on JanitorAI, then click on the “using janitor” button on the menu bar at the top to access the API settings menu.
Click on the “Proxy” option in the API settings menu and then on “+ Add Configuration.” There are three key options for configuring OpenRouter on JanitorAI.
- Model: Select the LLM model you want to use. You must enter the exact model name provided by OpenRouter.
- Proxy URL:
https://openrouter.ai/api/v1/chat/completions - API Key: Enter the API key you created on OpenRouter’s site.
Free DeepSeek Models On OpenRouter
Below are the free DeepSeek models available on OpenRouter recommended for AI roleplay. Make sure to copy and paste these in the Proxy API settings menu on JanitorAI.
tngtech/tng-r1t-chimera:free(Read our TNG R1T Chimera model feature article).tngtech/deepseek-r1t-chimera:freetngtech/deepseek-r1t2-chimera:free
Make sure there are no unwanted spaces or typos in your input. It’s best to copy and paste these details instead of typing them manually. Save your settings and refresh your chat page. You should now be ready to use the LLM model from OpenRouter on JanitorAI.
Use OpenRouter As A Proxy On JanitorAI For Other Models
You can use OpenRouter as a proxy on JanitorAI, allowing you access to other free or paid models.
Read Further: Use OpenRouter On JanitorAI
Errors After Configuring A Proxy On JanitorAI
Errors can pop up if you make a mistake while configuring a proxy on JanitorAI or if the DeepSeek API experiences a service interruption or outage.
Double Check Before Troubleshooting:
- Your API/Proxy URL is correct, with no extra spaces or typos.
- You clicked “Save Settings” and then refreshed your chat page.
- Your API key is correct, with no extra spaces or typos.
