Topaz Video Upscale
What is Topaz Video Upscale?
Topaz Video Upscale is the API form of Topaz Video AI's one-click upscaling. Send a video_url, optionally pick an upscale factor (1.25, 1.5, 2, 3, or 4) and a target_fps, and the model returns a higher-resolution video with optional frame rate retiming. The default factor is 2x, the maximum output is 4K (3840x2160), and source FPS is preserved unless you set target_fps explicitly. There is no enhancement-model selector or per-frame parameter to tune. The API is the production engine wrapped in a single call, which makes it the right surface for batch jobs and pipeline integration. Hand-tuned masters still belong in Topaz Video AI desktop.
Key features of Topaz Video Upscale
Four features cover what you'll set on a typical job.
Five upscale factors up to 4K
1.25x, 1.5x, 2x (default), 3x, and 4x. The maximum output is 3840x2160; sources already at or above that are rejected. Pick 2x for SD-to-1080p conversions and 4x for 1080p-to-4K masters.
Source FPS preserved by default
Leave `target_fps` unset and the output keeps the source frame rate. Set it to an integer (24, 25, 30, 45, 50, 60) and the output is retimed during the same upscale pass.
One-call upscale and retiming
Up-resolution and frame-rate change happen in the same job. No separate retime step, no second pass through a desktop tool, no intermediate file to manage.
Pipeline-friendly task shape
Same `POST /v1/tasks` shape as the rest of the platform. Submit a job, get a `task_id`, poll or wait for a webhook. Easy to drop into an existing batch processor.
Best for
Archive remasters
Push SD or HD masters to 2x or 4x for 1080p or 4K delivery files in a single call. No enhancement-model decision to make per clip.
Social repurposing
Resample legacy clips to a higher-resolution canvas before re-encoding for short-form social. Pair with a vertical reframe in your NLE.
Frame rate conversion
Retime 24 fps source to 60 fps for sports, gaming, or motion-heavy edits without a second tool in the chain.
B-roll enhancement
Boost detail without micromanaging enhancement-model parameters. Useful for atmospheric filler that needs to match a higher-res hero.
Pitch and proof-of-concept
One-click upscales when you want clean output without picking a preset. Good for client review rounds before committing to a desktop master.
Batch normalization
Drive the same factor across an asset library so the whole catalogue lands at consistent delivery resolution.
Use cases
Push an SD or HD archive to 1080p or 4K with a single 2x or 4x call per clip, then drop the results back into the catalogue. Convert 24 fps source to 60 fps for a sports edit by setting target_fps to 60 alongside an upscale factor; the retiming and the up-resolution happen in one job. Normalise a backlog of social cuts to a consistent delivery resolution by running every clip through the same factor. Stand up a one-click upscale step inside a content tool where users upload legacy footage and get a higher-resolution master back without picking a preset.
Limitations
The 4K output cap (3840x2160) means sources already at or above 4K cannot be upscaled further; the request returns an error. There is no enhancement-model selector, motion deblur control, or grain control like the desktop tool exposes. Output retains the source aspect ratio: if you need a different aspect, reframe in your NLE before or after the call.
API examples
Call Topaz Video Upscale from any language by POSTing to /v1/tasks. Full parameter docs live at docs.unifically.com/models/video/topaz-labs/video-upscale.
curl -X POST https://api.unifically.com/v1/tasks \
-H "Content-Type: application/json" \
-H "Authorization: Bearer YOUR_API_KEY" \
-d '{
"model": "topaz-labs/video-upscale",
"input": {
"video_url": "https://example.com/video.mp4",
"upscale_factor": 2
}
}'
Successful submission returns a task_id. Poll GET /v1/tasks/<task_id> or set a callback_url on the request to receive the finished video URL.
FAQs
People also ask
Topaz Video Upscale is the API form of Topaz Video AI's one-click upscaling pass. Send a video_url, optionally pick an upscale factor (1.25, 1.5, 2, 3, or 4) and a target FPS, and the model returns a higher-resolution video with optional frame rate retiming.
Leave target_fps unset. The model defaults to the source video's FPS, so the output preserves whatever frame rate the input was shot at.
Five. 1.25x, 1.5x, 2x (the default), 3x, and 4x. The maximum output is 4K (3840x2160). If the source is already at or above 4K, no factor can be applied and the request returns an error.
Same engine family. The API exposes a single one-click upscale plus optional FPS retiming. Topaz Video AI desktop adds enhancement-model selection, motion deblur, and other granular controls. Use the API for batch automation; use desktop for hand-tuned masters.
Yes. Pass target_fps as an integer (for example 30 or 60) and the output is retimed to that frame rate. Useful for converting 24 fps source to 60 fps for sports or motion-heavy edits.
A required video_url for the source clip, plus optional upscale_factor and target_fps. The output is delivered as a downloadable video URL once the task completes.
Per finished output second on topaz-labs/video-upscale. See the pricing page for the current rate.
