Contact
docs.typingmind.com
feedback@typingmind.com
support@typingmind.com
- Please email hello@lemonsqueezy.com with your receipt number.
- Company name: Devuap LLC
- Address: Ste 1200, 1309 Coffeen Avenue, Sheridan, WY, Sheridan, US, 82801
General FAQs
Frequently asked questions
Self-host FAQs
Frequently asked questions
Custom Deployment FAQs
Typing Mind Custom Deployment is a custom-branded, cloud-hosted version of Typing Mind. You can create and customize a Chat instance that works exactly like TypingMind.com under your own domain. You can offer the chat interface to your team members, community, or customers.
There are two sets of features. The Users Features (for your users): complete fully-featured chat experience like what you see on TypingMind.com. Admin Features: currently you can only customize a few settings and branding options. All the available customizable settings are listed in the Admin Panel. We'll work on adding more customizations soon.
Login to the Admin Panel and go to the Domains section. Add your domain and follow the instructions.
Please check if you've added the correct DNS records. You can find the DNS records in the Domains section of the Admin Panel. If you're still facing issues, please contact us
The Admin Panel is available at "/admin" of your Chat instance URL. You can also click the link "Admin" in the footer section of your chat UI's the sidebar.
Login to the Admin Panel and go to the Members section. Click on the "Add Members" button and fill in the details.
The default subscription comes with 5 users included. You can add more users at with a one-time purchase payment of $39 per user.
The early access price is a fixed $79/month subscription with 5 users included and 100,000 messages per month. You can add more users at with a one-time purchase payment of $39 per user. You need to provide your own Open API key and pay the API usage on your own. To increase the messages limit per month, please contact us.
The default subscription comes with 100,000 messages per month per instance. This limit is set to prevent abuse of our proxy server and hurt our bandwidth usage. If you reach this limit, please contact us.
A "Chat instance" or an "instance" is a chat interface you created when sign up for a Custom Deployment on TypingMind. You can create multiple chat instances for multiple purposes. Each Chat instance is tied to one OpenAI key and run on one domain.
Yes. After you create a Chat instance, you will need to add the OpenAI API Key in your admin panel. Otherwise, your users cannot chat.
No. Your users do not need an OpenAI API key to use your chat instance.
Your API key is stored securely in our database and is not shared with anyone. When your users send messages using your Chat instance, the requests are sent to our server, then we will use your OpenAI API key to get the response from ChatGPT. Your users will not see your API key.
No. After you add them to your Members list (in the Admin Panel), they can just login and use your Chat instance.
No. Your users will not see your API key. Requests to OpenAI API are sent from our server.
Currently, no. But we're actively working on making it possible. Stay tuned!
Yes. You can remove your users from the Members list in the Admin Panel.
Currently, we are not offering annual subscription. You can only subscribe monthly.
At the moment, you can customize the logo, tagline, and social poster image. Some more settings are available in the Admin panel. We'll make it possible to customize the interface more in the future.
Yes, but not yet. We're working hard to make all of these configurable in the admin panel. Make sure to check the Custom Deployment page at https://typingmind.com/custom and the Admin panel to see the latest feature updates!
Your chat interface will have exactly the same features as on TypingMind.com. Updates, new features, and bug fixes on TypingMind.com are also available on your chat instance at the same time.
Yes you can, however, at the moment you'll need to manage the user payments by yourself. You will also need to add your users manually via the Admin Panel. We'll look into making this easier soon.
The chats for each user are saved locally on their devices. As an Admin, you can enable chat logging (at and additional cost) that will allow you to view your users' chat messages. Note that the users will be informed that their chats are saved and can be viewed by you (the Admin).
Use the "Model & Training" page in the admin panel to train the AI chat with your data. The training is done by providing the chatbot with an initial system instruction, where you will put all relevant data to give the chatbot a context. More examples on how to train the chatbot is available in the "Model & Training" page in the admin panel
Yes. Tranining the model with your own data is currently done via the default system prompt with text input.
Training is currently made possible via the system instruction prompt. So the amount of data you can train depends on the context length of the model you have access to. GPT-3.5 has 4,096 tokens context length, GPT-4 has up to 32K tokens of context length. Keep in mind that you'll have to pay for your own OpenAI API usage, the more token used, the more costy it will get. In the future, we will look into utilizing ChatGPT Plugins to enable even more tranining data and knowledge-base based Q&A.
Yes, but not yet. We're working on this and will be available soon. Stay tuned!
Not yet, as ChatGPT plugins are not supported in the API. But once OpenAI makes it possible, we'll integrate it in right away.
Yes. We allow you to convert your Team License to the Custom Deployment version with the same amount of users. You will still have to pay the fixed subscription cost. Note that after you convert the Team License, the License Key will no longer work on TypingMind.com and your users/team will have to switch to using your chat instance instead. To convert your Team License, please contact us at support@typingmind.com.
The main different is that in the Custom Deployment, your users don't need to enter an OpenAI API Key and License Key, their chats will use your API key and you will pay for your users' usage. In the self-host static app (and the custom branding addon), you simply host a 1:1 copy of TypingMind, and your users still have to buy/enter a License Key on their own.
By default, yes. But you can hide it from the users and use your own default system propmt. This feature will be available soon.
For now, we support all Chat models from OpenAI, including GPT-3.5 and GPT-4. Fine-tune models are not yet supported.
No, the self-host static app that comes for free when you buy a license is not intended to be used by team. You can still allow your team members to use it but they'll need to enter their License Key and API Key. With the Custom Deployment, your team members do not have to have a license key and API key to use it.
Currently, that's not possible. For now, your users can only login via email.
No, that's not the intended use case for Custom Deployment.
Yes. That's doable by customizing the default system instruction and hide it from the users. This feature will be available soon.
That's not possible at the moment, but we're looking to add it in soon.
Terms of Service: https://typingmind.com/terms#custom-deployment ; Privacy Policy: https://typingmind.com/privacy#custom-deployment
It's one of our new product offering, but it's not ready yet. Stay tuned!
Please send an email to support@typingmind.com
