Iterate cheap, deliver big
Generate at 480p or 720p while you are still testing motion, then upscale only the take you kept. You stop paying high-resolution render time on clips that get discarded.
Upscale Video enlarges a clip up to 4x using Topaz Labs enhancement models, with separate models for general work, denoise, faces, motion deblur, CG, and old footage. Frame interpolation is available on the same node for FPS conversion and slow motion.
In Phygital+ it sits at the end of a video chain: generate at a lower resolution while you are iterating, then push the take you kept up to delivery size, in the container the next step expects.
Delivery-resolution exports, republishing old footage, FPS conversion, and slow motion
Video, social, and archive teams preparing clips for a bigger screen or a different spec
Up to 4x enlargement with per-content enhancement models and optional motion control.
Generate at 480p or 720p while you are still testing motion, then upscale only the take you kept. You stop paying high-resolution render time on clips that get discarded.
Proteus for general work, Iris for faces, Gaia for CG and generated animation, Themis for motion deblur, Dione for old footage. Matching the model to the content beats reaching for a stronger general setting.
Raise the frame rate, smooth out motion, or produce slow motion at up to 16x, all on the same node. It stays off by default, because interpolation changes how motion reads and that is not always wanted.
Old, low-resolution, or heavily compressed clips get a second life when the source still holds usable detail. Republishing archive footage stops meaning a reshoot.
Pick mp4, mov, or mkv and set crop-to-fit behaviour so the clip comes out matching the delivery spec. No extra conversion pass between the upscale and the edit.
Upscale Video fits at the end of a workflow, where a clip has to reach delivery resolution or an old file has to survive a modern screen.
Connect a clip, pick a scale and enhancement model, and run.
Connect the source clip to Your Video. Start from the cleanest version you have, since upscaling restores structure rather than inventing detail that was never captured.
Pick X2 for balanced runs and X4 only when the source supports it. Keep Use Upscale enabled and match the enhancement model to the content: Proteus for general work, Iris for faces, Gaia for CG, Themis for motion blur, Dione for old footage.
Turn on frame interpolation only if you need smoother motion, a different frame rate, or slow motion. Set the output container and crop-to-fit behaviour for your delivery spec, then run and review.
One subscription across 30+ AI models, with no per-tool credit balances or separate signups. Credit cost per run is shown live in the node before you start it.
500 weekly credits to test the node and see the output quality. No credit card required.
45,000 monthly credits with cheaper per-credit pricing, commercial rights, and video download.
90,000 monthly credits, up to 10 seats, shared workspace, centralized billing, and priority support.
210,000 monthly credits, API and integration support, unlimited seats, and dedicated management.
Full specs for Upscale Video in Phygital+.
How Upscale Video compares with the other enhancement nodes in Phygital+.
Put video upscaling at the end of a generation chain instead of running it as a separate tool.
Generate at 480p or 720p to keep iteration cheap, then upscale only the take you kept.
Add narration, dialogue, or sound design to the upscaled clip in an audio node.
Upscale a generated still first, then use it as the start frame so the clip begins from the sharpest version.
Restore an old clip, then extend or restyle it in a video node that accepts a video input.
Everything you need to know about Upscale Video in Phygital+.
Upscale Video enlarges a source clip up to 4x using Topaz Labs enhancement models, with optional frame interpolation for smoother motion or slow motion. It also controls the output container and crop-to-fit behaviour, so a clip comes out ready for the next step rather than needing a conversion pass.
X1, X2, X3, or X4. X2 is the balanced choice and the right setting for review runs. Save X4 for cases where the source quality genuinely supports a larger final output, since upscaling cannot recover detail the original file never captured.
Proteus is the general-purpose default. Artemis handles denoise and sharpen, Nyx targets noise, Gaia suits CG and generated animation, Iris is built for faces, Themis handles motion deblur, and Dione is for old content. Match the model to the content type rather than assuming a stronger setting is better.
It generates intermediate frames, which lets you raise the frame rate, smooth out motion, or produce slow motion at up to 16x. It is off by default and should stay off unless you specifically need one of those things, because interpolation changes motion rather than just resolution.
Set the FPS target and, for slow motion, the slow motion rate. Leaving FPS at 0 uses the source frame rate behaviour where supported. If interpolation introduces artefacts, set the interpolation model back to NULL or turn the whole feature off.
mp4, mov, or mkv. Use mp4 for general delivery and broad compatibility, and pick mov or mkv only when the next tool in your pipeline expects that container.
Often, yes, if the source still contains usable detail. Dione is built for older content and the enhancement pass can meaningfully clean up low-resolution or heavily compressed footage. What it cannot do is invent detail that was never recorded.
Usage is credit-based and included in every plan, from Free up to Enterprise. The credit cost per run is shown in the node before you start it, so there is no separate Topaz licence to buy.
Outputs created on any paid plan come with commercial-use rights, including clips you upscaled from your own footage.
Browse the full catalog of 30+ AI models available in Phygital+.
Join 100+ teams using Phygital+ – every model in one workspace
Try Phygital+ free