Usage and Limits

Restrict model and resource access for your members

In the quest to optimize the use of TypingMind custom as an AI chat app for your team, it's essential to have control over model usage limits and resource avaibility for your members.

Now you can limit user access to certain elements:

• Specific prompts
• AI Agents
• Chat models

This can be achieved through our member tagging system:

• Assign tags to your team members to organize them into different groups.
• Apply these tags to particular resources or chat models to control access according to those groups.

Let’s check out the details!

Set Tags for your team members

To control the user access to any resources, you need to set up tags for them first. Here’s how:

• Go to Members
• Click on a member
• Select or update a new tag in the Tags section
• Enter a new tag (for example: “marketing”, “dev”, "content")

Please note that you can add more than 1 tag for your members.

Restrict model and resource access with member tags

After completing your setup with member tags, please follow the steps below to restrict member usage to specific chat models, prompts, or AI Agents.

1. Restrict chat model usage

You can restrict the model usage of your team members in the following settings:

Control model availability

Model availability helps you control which AI models are accessible to which user tags.

To set up model availability for certain user groups, you'll need to:

1. Navigate to the "Usage & Limit" section in the Admin Panel.
2. Choose the AI model.
3. Set the availability to "Available to users with tags" and enter the relevant tags.

Example: If you want only the Marketing Team to use GPT-4 for tasks like content generation, simply assign a "marketing" tag for the GPT-4 model availability.

Limit the number of messages that can be sent

Message limits allow you to control the number of messages that can be sent (by specific user groups or all users) within a certain time frame.

To set up message limits:

1. Go to the "Usage & Limit" section in the Admin Panel.
2. Choose an AI model.
3. Click "Add Limit" at Message Limits
4. Set a limit on the number of messages per (x) hour
5. If this applies for certain users, then add tags of that user to "Apply for users with tags"

For example, you might allow the Content and Technical team to access Claude but set character limits for each group:

• 50 messages/hour for “content” tag
• 25 messages /hour for “technical” tag

Limit the number of characters that can be entered per message

This allows you to restrict the number of characters that can be used in each message sent (by specific user groups or all users).

To set up character limits:

1. Go to the "Usage & Limit" section in the Admin Panel.
2. Choose an AI model.
3. Click "Add Limit" at Character Limits
4. Set a limit on the number of characters per message
5. If this applies to certain users, then add tags of that user into "Apply for users with tags"

For example, if you have partners who you've invited as "guests", you might restrict them to using GPT-4o with a limit of 100 characters per message.

Limit the number of characters that can be sent per time period

This allows you to restrict the number of characters that can be sent in a specific time period by specific groups of users.

1. Go to the "Usage & Limit" section in the Admin Panel.
2. Choose an AI model.
3. Click "Add Limit" at Character Limits Per Time Period
4. Set a limit on the number of characters every x hour/days/months
5. If this applies to certain users, then add tags of that user into "Apply for users with tags"

Check other ways to control the chat models.

2. Restrict Prompt access

To assign a member tag to a prompt, do as follows:

• Go to the “Prompts Library” section from the Admin Dashboard.
• Select “Add Prompt” to create a new prompt, or “Edit” an existing prompt.

• Scroll down to the “Visibility” section and select from the drop-down menu:
  • “Visible only to users with tags”: once you add tags, users in these certain tags are allowed to use the model
  • or “Visible to all users except users with tags”: once you add tags, users in these certain tags ARE NOT allowed to use the model.

For example, if you want only members with the "content" tag to access the “English pronunciation” prompt, simply add the "content" tag to that prompt.

Here's what the chat interface will look like for team members with the "content" tag:

3. Restrict AI Agents access

To assign a member tag to a prompt, do as follows:

• Go to the “AI Agents” section from the Admin Dashboard.
• Select “Create AI Agent” to create a new prompt, or “Edit” an existing prompt.

• Scroll down to the “Visibility” section and select one of two options from the drop-down menu.
  • “Visible only to users with tags”: once you add tags, users in these certain tags are allowed to use the model
  • or “Visible to all users except users with tags”: once you add tags, users in these certain tags ARE NOT allowed to use the model.

For example, if you want only the team members with the "content" tag to access the “Marketing Expert” agent, simply select the "content" tag for that AI Agent

Here's what the chat interface will look like for team members with the "content" tag:

Conclusion

The examples above make it easy to understand how to manage access restrictions with TypingMind Custom. This approach is versatile and can be used for:

• Separate access among different teams or departments in your organization.
• Customize access for different groups of customers.
• Adjust access for users based on the packages or tiers you define.