tt by teramoby
/ Guide
EN 中文

Guide · Updated August 18, 2026

How to Configure tt and Start a Multi-Model Conversation

tt is a native multi-model AI client for iPhone, iPad, Mac, and Android. This guide shows how to connect your AI services, ask your first question, and read the result.

In a standard chat, tt sends the same question to every service whose model and credential configuration is complete, preserves each original response, and asks your chosen summarizer to organize the common points and differences into one synthesized answer. With one configured service, tt shows that service's response directly.

Team Mode is a different workflow: participating models can read the shared discussion, respond to one another, and stop when they have nothing useful to add. In either mode, a model-generated answer can still be wrong. For important decisions, inspect the original responses and verify claims against reliable sources.

Installation

Open the direct App Store link on this site, or search for teramoby and download “teramoby tera (tt)” for free. On first launch, review the privacy disclosure and continue into the app.

⚠️ Note: tt is not currently available on the China App Store due to compliance requirements.
tt by teramoby on the App Store
Search for teramoby on the App Store to find teramoby tera (tt)

Setting Up API Keys

tt is a client — it doesn't supply AI of its own. Model requests travel directly from your device to the configured providers. When Online Search is active and a search is needed, a derived query also goes to the search endpoint described in the Privacy Policy. To get started, enter an API key from at least one provider in Settings → Providers. Here are official links for several common providers:

See the current AI provider and hosting-platform integrations. Exact configuration choices appear in the app.

Configure two or more providers to compare more perspectives. Model count alone does not make an answer correct; review the originals and verify important claims.
tt multi-model chat home screen
The main screen, ready for your first question
tt AI provider and API key settings
Settings → Providers: configure your LLM providers and API keys

Asking Your First Question

Once your keys are in, you're ready. The workflow is simple:

Check your configured services

Every service with a complete model and credential configuration will participate in the chat workflow.

Type your question and send

tt sends the request to all completely configured services in parallel.

Read each model's response

Responses come back simultaneously. You can read each model's full answer individually.

Review the synthesized answer

tt sends completed responses to the summarizer model, which organizes them into one answer. You can still return to every original response.

tt synthesized multi-model answer
The Summarizer model synthesizes all responses into one final answer

Questions or feedback? Email tera-support@teramoby.com or join the Google Group.