-
Notifications
You must be signed in to change notification settings - Fork 0
Gen3 Model Guides
GT AI OS Release edited this page Jun 18, 2026
·
3 revisions
- Open GT AI OS Instructions and expand Model Guides in the sidebar.
- Pick the guide for the model or capability you are configuring (for example Orpheus v1 English — Vocal Directions for expressive text-to-speech in GT Chat).
- Confirm the model appears in your tenant Model Catalog and that operators enabled the matching capability on Control Panel Models.
- Apply agent-level settings on Building Agents when the guide references Speech, Vision, or Image tabs.
Model guides explain how to use a specific inference model inside GT AI OS—not just that the model exists in the catalog. They cover agent configuration, chat affordances (for example Read aloud), and provider-specific syntax (such as Orpheus vocal directions) so you get the behavior you expect without guessing.
Vendor pricing, rate limits, and raw API parameters live on provider model cards; see Model Cards for official Groq and Google Gemini documentation links.
| Guide | Model | Use when |
|---|---|---|
| Orpheus v1 English — Vocal Directions |
canopylabs/orpheus-v1-english (Groq) |
You want expressive TTS in GT Chat with bracketed directions like [cheerful] or [whisper]
|
- Model Catalog — browse enabled models in your tenant
- Model Cards — official Groq and Gemini model documentation
- Accessibility — STT/TTS prerequisites and Read aloud troubleshooting
- GT Chat — message actions including Read aloud
- Building Agents — Speech Settings tab