Skip to content

Cookie preferences

We use cookies for our own analytics, to see how our campaigns perform and, if you allow it, to load content from other services such as the Google site search. We never sell your data. Necessary cookies keep the site working and cannot be switched off. See our Cookie Policy for details and our Privacy Policy for how we handle personal data.

Security, load balancing and remembering your cookie choice.

Show us which pages people read, how they find the site and where they get stuck, so we can improve it (Google Analytics, Microsoft Clarity).

Show us which of our ad campaigns bring visitors to the site and remember the campaign or partner link you arrived from (Google Ads, LinkedIn, Meta, partner program). We do not use these cookies to build advertising profiles.

Load the site search from Google when you use it. Google sets its own cookies and shows ads in the search results.
© 2026 The ThingsBoard Authors
22.6k Try for free

ThingsBoard Cloud

Choose your data region

Your data stays in the region you choose, for residency and compliance. No credit card required.

North America Data stored in North America Europe Data stored in the European Union

Rather run it yourself? Install on your own servers

AI Settings

Trendz AI features send requests to a large language model (LLM) that you connect in Settings > AI Settings. This page shows how to add a model, check that Trendz can reach it, and choose which features use it.

AI feature Where you use it
Prompts AI Widget Summary and other user-created prompts
View Assistant The AI Assistant chat that answers questions and builds views
Metric Code Assistant Metric generation in Metric Explorer

Before you start

Section titled “Before you start”

Get credentials from one of the supported providers:

Provider What you need Where to get it
Google Gemini (recommended) API key Google AI Studio
OpenAI API key OpenAI platform
Amazon Bedrock AWS region, access key, and secret key AWS Bedrock API guide

Open AI Settings

Section titled “Open AI Settings”
  1. Click Settings (gear icon) in the bottom-left corner of the left menu.
  2. Open the AI Settings tab.
Trendz home page with the Settings gear icon in the left menu highlighted Settings page with the AI Settings tab highlighted

Add an AI model

Section titled “Add an AI model”
  1. On the AI Settings tab, click New AI Model.
  2. Enter a Name that identifies the model in Trendz, and choose the Provider.
  3. Choose the Model Name. Trendz fills in the provider’s newest chat model, and the list under the field suggests the others, newest first. Type to filter it, or enter any other model name the provider accepts. The suggestions come from the public models.dev catalog: if Trendz cannot reach it, the field stays empty and you type the model name.
  4. Enter the credentials. For Google Gemini, that is the API Key.
AI Settings tab with the New AI Model button highlighted New AI Model dialog with the provider list open: OpenAI, Google Gemini, and Amazon Bedrock Model Name field with gemini-3.5 typed and the matching suggestions highlighted API Key field filled in and highlighted
  1. Click Check connectivity. Trendz sends a test request to the provider with the values in the form.
  2. Wait for the result. Connected means the model is ready. On Not connected, hover over the message to see the provider’s error, fix the field it names, and check again.
  3. Click Save. The model appears in the AI Models list.
New AI Model dialog with the Check connectivity button highlighted Model dialog showing Connected after the check, highlighted New AI Model dialog with the Save button highlighted AI Models list with the new Gemini Flash Lite model highlighted

You can add several models, like a fast and inexpensive one for prompts and a more capable one for metric code.

Turn on AI features

Section titled “Turn on AI features”
  1. Turn on Enable AI Features.
  2. Choose how features get their model:
    • Use Single Model on: select the model in Single AI Model. Every feature uses it.
    • Use Single Model off: select a model for Prompts, View Assistant, and Metric Code Assistant separately. Select Disabled to turn a feature off for everyone in the tenant.
  3. Click Save.
Enable AI Features toggle highlighted Single AI Model field highlighted with Gemini Flash Lite selected Prompts, View Assistant, and Metric Code Assistant fields highlighted, each with a model selected Prompts list open with the Disabled option and the Gemini Flash Lite model highlighted Save button of the AI features section highlighted

Edit a model

Section titled “Edit a model”

Change a model’s name, provider, model name, or credentials, like when you rotate an API key or move to a newer model. Features that use the model switch to the new values as soon as you save.

  1. In the AI Models list, click the actions menu (…) in the model’s row and select Edit. Clicking the row itself opens the same form.
  2. Change the fields you need. The form is the same as New AI Model, and its title shows the model you edit.
  3. Click Check connectivity and wait for Connected, especially after you change the key or the model name.
  4. Click Save. Trendz shows AI Model was saved, and the list shows the new values.
Actions menu of the Gemini Flash Lite row open with Edit highlighted Edit Gemini Flash Lite dialog with the Model Name field changed to gemini-3.5-flash and highlighted Edit Gemini Flash Lite dialog with the Save button highlighted AI Models list with the updated row highlighted and the AI Model was saved message at the bottom

Delete a model

Section titled “Delete a model”
  1. Make sure no AI feature still uses the model. If one does, select another model for it in the AI features section and click Save.
  2. In the AI Models list, click the actions menu (…) in the model’s row and select Delete.
  3. Click Yes to confirm. Trendz shows Success!, and the model disappears from the list.
Actions menu of the model row open with Delete highlighted Confirmation dialog asking Do you confirm the deletion of this model, with the Yes button highlighted Empty AI Models list with the Success message highlighted

Provider fields

Section titled “Provider fields”

Every model has a Name, which only labels it in Trendz, and a Provider. The provider decides which other fields the form shows.

Provider Field What to enter
Google Gemini Model Name The Gemini model ID, like gemini-3.5-flash-lite
API Key An API key from Google AI Studio
OpenAI Base URL The API endpoint. Leave the default https://api.openai.com/v1 for OpenAI itself, or enter the endpoint of any service with an OpenAI-compatible API, like a self-hosted model server
Model Name The OpenAI model ID
API Key A secret key from the OpenAI platform
Amazon Bedrock Region The AWS region where Bedrock gives your account access to the model, like us-east-1
Model Name The Bedrock model ID or inference profile ID, as listed in the Bedrock console
Access Key The access key ID of an IAM user allowed to invoke Bedrock models
Secret Key The secret access key of that IAM user. See the AWS Bedrock API guide

When you choose a provider, Trendz fills Model Name with that provider’s newest chat model. Replace it if you want another model. Keys show as dots in the form, including when you edit a saved model.

Token prices depend on the model and change over time. Check the provider’s pricing page before you choose a model.

Google Gemini fields highlighted: Model Name and API Key OpenAI fields highlighted: Base URL, Model Name, and API Key Amazon Bedrock fields highlighted: Region, Model Name, Access Key, and Secret Key
© 2026 The ThingsBoard Authors · Privacy Policy · Cookie Policy ·