Recently, I discovered Janitor AI, It is a fantastic AI tool that lets you create fictional chatbot characters with different personalities. It has fictional characters of Spider-Man, Gwen Stacy, or your favorite game characters like Leon and Ghost and you can literally talk to them, all with the help of generative AI.
What’s more, if you’re a developer, you can integrate these chatbots into your own apps. Let’s see how you can use Janitor AI for free and start creating your own interactive experiences.
What is Janitor AI
Janitor AI is a platform where you can interact or chat with ai chatbots of fictional characters. It was launched in June 2023 and has really gained decent popularity, with over a million people jumping on board in just a month!
You can choose from a huge collection of chatbots made by other users, join in on role-playing adventures, or even get creative and design your own characters with unique personalities and dialogues. The best part? You can also plug these chatbots into your own apps using APIs and SDKs, which is a big plus for developers out there.
If you love chatbots, Janitor AI has a fantastic community that will help you at every step. Hang out on their Discord, join events, showcase your creation, and get the needed support. This is a real fun and friendly place to connect and share!
Is Janitor AI Free to Use?
Yes, Janitor AI is completely free to use. Simply create an account on their website to access various chatbot personas and features. While basic usage is free but if you are a developer and want to integrate it into your apps then there is third party api pricing involved i.e. openai api.
Janitor AI API Settings and Price for developers
For developers looking to integrate Janitor AI into their applications, understanding the API settings and pricing is crucial.
To use Janitor AI in your apps, you need to integrate any of these AI API such as Janitor LLM, OpenAI or KoboldAI. Here’s a brief overview of the setup:
1. OpenAI API:
- Create an API key: Sign up on the OpenAI platform and generate an API key.
- Configure the API in Janitor AI: Enter the API key in the Janitor AI settings.
- Monitor Usage: Keep an eye on your API usage to avoid unexpected costs.
2. Kobold AI:
Local Setup: KoboldAI can be run locally if you have a powerful enough PC.
Installing Kobold AI API for Janitor AI:
If you’re opting for Kobold AI in your API settings, ensure you’re using the KoboldAI United version. You also have the option to rent a GPT for $0.2 per hour.
Follow these steps to install KoboldAI United:
- Download and Install KoboldAI: Go to the KoboldAI GitHub page, download the ZIP file, and follow the installation instructions. Make sure to run the installer as an Administrator.
- Set Up Pygmalion 6B Model: Open KoboldAI and confirm it’s running correctly. Install the Pygmalion 6B model. Check your GPU’s VRAM and allocate resources appropriately between your GPU and CPU.
- Connect to JanitorAI: Launch the ‘remote-play’ batch file, copy the link it generates, and paste it into the JanitorAI API URL box. Save your settings and start using the service.”
Hosted Solutions: Alternatively, you can use hosting services like Mancer or OpenRouter for larger models.
Integration: Similar to OpenAI, you’ll need to configure the API settings within Janitor AI using the provided URL from the hosting service or your local setup.
Pricing:
Janitor AI offers different pricing models based on usage and features:
- Free Plan: Access basic premade characters at no cost. Users are responsible for the costs associated with the AI APIs (e.g., OpenAI or KoboldAI).
- AI API Costs:
- OpenAI: Comes with $5 free credit, after which it costs depending on usage.
- KoboldAI: Typically costs $0.20 per hour for hosting services only.
These plans and settings provide flexibility for both casual users and developers looking to integrate janitor AI character bots into their applications.
Read More : Profile Pic Maker AI (PFPMaker.ai)
How to use Janitor AI?
Here is a brief guide to making use of the janitor ai:
Getting Started with Janitor AI
Sign Up or Log In: Go to Janitor AI and sign up using your email to sign up, or log in if you already have an account.
Browse Available Chatbots
Browse available chatbots and virtual assistants. Typically, you will find chatbots for various purposes—from personal assistants to customer service, entertainment, and many others.
Select a Chatbot
Select any chatbot of your choice. Each chatbot will be provided with a description and ability to perform tasks.
Chat with the Chatbot
Start chatting with the chatbot by typing any question or command in the text box that is available to you. The reply that you receive from the chatbot is based on their settings and AI learning.
Creating a Chatbots (Optional)
platform also allows you to create your own chatbots. This includes:
1. Setting Up Responses
- Define how your chatbot should respond to specific queries.
2. Integrating APIs
- Connect your chatbot with external APIs to add advanced functionalities.
3. Training the AI
- Provide data or scenarios to train AI to improve performance.
Deployment and Usage
Once the chatbot has been chosen and developed according to the specified requirements, it can be deployed on websites, inside applications, or for personal use.
Common Issues: Why Is Janitor AI Not Working?
I am listing down some very common errors that people are facing and searching for in online communities like Reddit. I got all these possible errors from these communities and now providing the impactful solutions here, which you can give a try.
1. Failed to Fetch / Network Error / Load Failed Error
These are the errors you see due to network issues with Janitor AI servers, your internet connection, or browser settings. You will receive error messages that might look like:
- “Failed to fetch”
- “Network error.”
Here are the recommended solutions:
- Check Janitor AI’s Discord for any issue with the server.
- Clear your browser’s cache and cookies.
- Use a VPN.
- Create a new API key.
- Wait for the team to fix server problems.
To avoid this, clear your cache and update your browser.
2. Chat Freezing Error
This is where the chat interface becomes unresponsive—more than likely because of server overload but could be due to many client-side issues. An error message may not be presented, just a chat that isn’t responding.
Solutions:
- Refresh the page.
- Use another device or browser—clear cache and cookies.
- Recheck your internet connection.
Prevent this by limiting open browser tabs and applications.
Read More : MaxAI.me Extension: Features, Pricing, use cases & More
3. Server is Busy Error
It means high traffic volumes are coming in on Janitor AI’s servers at that very time that’s causing your request not to be processed. You may encounter:
- “Server is busy”
- “High traffic; please try again later.”
Solutions:
- Wait a few minutes and retry; do so during off-peak hours; follow Janitor AI on social media for updates.
Mitigate this by working less busy hours or upgrading to a premium account.
4. 403 Error / Proxy Error 500
The error code 403 relates to insufficient permission for the resource, while the proxy error 500 is shown in case of server misconfiguration.
You can work around this by checking settings for the API key verifying the API credits, reconfiguring the proxy, or contacting support. Get ahead of these by regularly auditing your use and settings for your APIs.
5. “No Active OpenAI Keys Available” or “You Reached Your Quota.”
These are thrown when either your OpenAI API key is misbehaving or, most likely, when you are running over the allowed usage for some time.
The solutions are to check for a valid API key and to generate a new key if you have to, as well as look into usage limits. You should monitor API usage and set up alerts as your usage approaches your limit to avoid this.
6. “No GPT-4 Keys Available.”
This error indicates that your account cannot access any GPT-4 models.
A solution could be to select a non-GPT-4 model, test your account type’s GPT-4 access, or contact support. Avoid this by keeping current on model access and account limitations.
7. “Max_length Should Be >1 And <= 512”
Configuration issue: the parameter contains the maximum length of the input or output text. Correct by setting max_length to a value in the range 1-512 and ensuring the value of the custom code inputs be <= this value.
Prevent by validating inputs and settings before requests are sent to the API.
8. “KoboldAI Ran Out of Memory: CUDA Out of Memory”
This error arises when KoboldAI runs out of memory on the GPU. Some possible workarounds include reducing the complexity of requests, closing all the other GPU-heavy apps, or upgrading to another machine with more excellent GPU memory. You can prevent this by monitoring your GPU usage and possibly upgrading your hardware in this respect.
Conclusion
Janitor AI offers a unique and engaging experience for fans of fictional characters and anyone who enjoys creative chatbots. Janitor AI has quickly become a popular platform due to its diverse library, role-playing options, and open development tools.
so if you want to chat with your favorite heroes or create your own unique characters? Janitor AI is the platform you looking for! With tons of existing chatbots and the ability to build your own, you can have endless conversations and role-playing adventures.
Plus, it’s free to join the fun and there’s a friendly community to connect with too. So why wait? Sign up for Janitor AI today and start creating your own stories!
