Studio
Images, video, speech and ComfyUI workflows in one project-aware studio
Studio generates images (text-to-image, image-to-image, inpainting, upscaling), video (text, image or audio to video, plus LTX retake and extend), speech and ComfyUI workflow outputs inside the active project. Image work runs on your own GPU through Forge or ComfyUI via TektBridge, locally or on a trusted laptop over P2P; video and cloud image models run on platform keys or your own BYOK keys. Results auto-save to the project and hand off to SyncCut, DocMan and Dashboard.
Who it is for: Engineers, architects and creators who need renders, textures, concept images, short clips and voice-overs next to their project documents.
Quick answers
Do I need a GPU?
Only for Forge and ComfyUI. Cloud image models, all video models and BYOK speech run on provider APIs; a trusted laptop's GPU can also be used from a phone over P2P.
Why is the model list empty?
Studio could not reach TektBridge on this device. Start the bridge with Forge, or pick a remote device in the Compute bar; the hosted server has no Forge of its own.
What are the starred models in the dropdowns?
Models you curated per Studio input in Settings → AI Models. They run on the key in your encrypted vault, which you unlock once in Settings.
Where do my images go?
Every Forge result auto-saves to generations/ in the project, explicit saves go to the project root or audio/, and Studio's state file keeps the last 50 generations. All of it syncs with the project.
Can I edit the result in another app?
Yes: SyncCut, DocMan and Dashboard buttons sit under every result, and img2img, img2vid, Retake, Extend and audio2vid chain tabs inside Studio.
Loading the user manual…