AI Settings

Configure AI Chat, AI fields, AI automation, and App Builder for a self-hosted instance.

Available for self-hosted Business plan and above

Path: Admin Panel → AI Settings

The AI Settings page is used to configure AI Chat, AI fields, AI automation, and App Builder. The Pending configuration panel on the right shows which required items are still missing.

Before you start

Before you open the AI Settings page, make sure the following items are ready.

Requirement	Required?	Notes
Agent Runtime setup	Required for AI Chat and App Builder	Used by AI Chat and App Builder
Custom provider	Required when AI Chat or App Builder uses your own key (BYOK)	Use an OpenAI Compatible or Anthropic provider
OpenAI API Key	Optional	Only needed if you want to enable Voice input in AI Chat
Public access	Required when the Vercel deployment engine is selected	Your Teable instance and object storage (MinIO / S3) must be publicly accessible to Vercel

Setup steps

Start with the AI Chat runtime, app deployment, and model setup:

1. Configure the AI Chat runtime and app deployment

   Choose a deployment option and complete the deployment, then click Run test and Test Connection.

2. Configure LLM API

   Set up your model connection and complete the initial test.

3. Configure Recommended Models

   Choose the recommended models for AI fields and AI automation.

4. Set chat model tiers

   Choose a model for each of the four chat model tiers.

Configure the runtime

Click Run test to verify the AI Chat runtime.

Configure LLM API

In LLM API, add a model provider and test its models. If you use your own key (BYOK), choose OpenAI Compatible or Anthropic as the provider type.

A model’s context window and maximum output per response are resolved automatically from the Model catalog; there is nothing to fill in. To check the values resolved for a model, open Model catalog, choose a Scope under Inspect model and select the model; Parameter details then lists the context, max output, and the source of each value.

When the catalog holds no matching entry (a newly released or self-built model, for example), the model runs on the values set under Nothing matched in Matching rules: a context window of 200000 and a max output of 32000 by default, with Matching rule shown as the source of both in Parameter details. If those values do not fit the model, adjust them in Matching rules and click Save.

Configure Recommended Models

Choose the recommended models users can select in AI fields and AI automation.

Set chat model tiers

In Chat model tiers, pick a model for Ultra, Smart, Standard, and Lite; the models you can pick are the ones connected in step 2. Users choose by tier in the model menu of sidebar AI Chat and see the tier name rather than the model name, so you can swap the model behind a tier at any time without affecting anyone who already picked it.

Tier	Description
Ultra	Top capability, for demanding work. Left empty it shows Not offered and stays out of the user’s model menu.
Smart	The main model. AI fields, background tasks, and any tier without its own model use it, so it must be set.
Standard	Everyday chat and general tasks; also used for medium background tasks. Left empty it falls back to Smart.
Lite	Low-cost, high-frequency routine work; also used for lightweight tasks such as generating chat titles. Left empty it falls back to Smart.

Each tier has two more controls on its right:

• Set as default: the tier users land on when they open the model menu. An Ultra with no model cannot be the default.
• Enabled: turn it off and the tier is no longer offered to users. The default tier cannot be turned off.

Enable AI features

In AI Capabilities, enable AI Field and AI Chat as needed. Configure the runtime before using AI Chat.

Configure App Builder

In Deployment Engine, choose how apps are deployed:

Deployment engine	Setup requirements
Vercel	Enter a Vercel access token, click Test Connection, and use Test Public Access to confirm that Vercel can reach your Teable instance and object storage.
Teable Infra	Uses the configured Teable Infra container runtime. No additional credentials or public access test is required here.

You can add these optional App Builder settings when needed:

Optional item	When you need it
Custom domain	When you want to publish apps under your own domain
App Auth	When generated apps need Email OTP or Google sign-in

In App Auth, you can configure sign-in providers for generated apps:

• Email OTP uses an independent SMTP mailbox for app login verification emails.
• Google OAuth requires a Client ID and Client Secret. Add the redirect URI shown on this page to the authorized redirect URI list in Google Cloud Console.

Teable sign-in is available by default. Email OTP appears in App Builder only after the SMTP settings are complete, and Google appears only after both OAuth credentials are configured.

Auto-inject AI API Key into apps is on by default, so apps can call the platform’s AI models. Turn it off and apps can no longer reach them.

Optional. Configure Voice input

To use voice input in AI Chat, enable the feature and enter an OpenAI API Key. Configure the Transcription Endpoint only when using a custom transcription service. You can also limit the recording length and requests per minute.

Verify the setup

After the setup is done, first confirm that all required items in the Pending configuration panel are green. Then test AI Chat, AI fields, AI automation, and App Builder one by one.

FAQ

I get a credit exceeded error

Check the account or provider used by your configured model service. Top up or change the model configuration if needed.