Skip to main content

Video Upscaler Review: 720p to 4K, Same Clip

Pick Video Upscaler to deliver a clip at a larger tier before publishing; generate at a size that renders fast, then upscale only the cut you keep.

Open the Playground

Spicy Index

56.3/ 100

Preliminary

Freedom Score

56.0/ 100

Range if the rest pass: 56.0–100.0

Content levels

Tested 3 of 6 probe types · 12 runs

  • Mild · The lightest prompts in the test

  • Moderate · Where filtered models start to refuse

  • Strong · Mid-scale; must be tested before a model is ranked

  • Very strong · Where most models start to hold back

  • Maximum · The top of the scale, scored in two halves: class A and class B

Price
—
Median time
1.7 min
Success rate
Collecting data

Video Upscaler takes one clip and returns the same clip at a larger resolution tier: 720p, 1080p (the default), 2k or 4k. Length, aspect ratio, motion, cuts and the audio track all come back unchanged — only the pixel size changes. video_url is a direct link to a file 3 to 60 seconds long. Pricing runs per second and scales with the tier: $0.006 at 720p or 1080p, $0.012 at 2K, $0.024 at 4K.

The verdict

The tier is a pixel budget rather than an exact target, so an 832×480 source comes back at 1896×1094 at the 1080p setting rather than an exact 1920×1080 — worth knowing before you build a delivery spec around an exact frame size. Like the image upscaler, this adds pixels to what's already there; it doesn't invent motion detail or sharpen footage that was soft or heavily compressed to begin with.

The natural pairing is a generation model first and this endpoint last: render a clip at a size and length that's fast and cheap to iterate on, pick the one cut worth keeping, then upscale only that clip before publishing. Running every draft through this step would cost more than it needs to; running the one you're actually shipping is where the per-second price is easy to justify.

Try Video Upscaler →

Key facts

  • Price: $0.006 per second at 720p or 1080p, $0.012 at 2K, $0.024 at 4K.
  • Input: video_url (a direct link, 3–60 seconds), resolution (720p, 1080p default, 2k or 4k). No prompt.
  • Output: one MP4 per request, the same length as the source, with motion, cuts and audio unchanged.
  • Sources: the parameter list and the published example live on the Video Upscaler model page.

What we liked and what held it back

What we liked

  • Everything except the pixel count stays put. Length, aspect ratio, motion, cuts and the audio track all come back as sent, so the upscaled file drops into the same edit or upload flow as the source did.
  • 720p and 1080p cost the same. Both sit at $0.006 per second, so there's no reason to pick 720p over 1080p on price alone if 1080p is the tier you actually need.
  • The published example is concrete about what changes. A 5-second 832×480 clip at the default 1080p tier returns at 1896×1094, with the climb, the lantern glow and the timing unchanged — resolution is the one thing that moved.
  • It scales to a real production clip length. 3 to 60 seconds covers most short-form delivery cuts without splitting a clip into pieces first.
  • The audio track survives the pass. A clip with sound comes back with that same sound, so this can sit after a generation or editing step without a separate audio re-attach.

What held it back

  • The tier is a budget, not an exact frame size. 1080p doesn't guarantee 1920×1080; the published example returns 1896×1094 from an 832×480 source, because the target is a pixel budget that follows the source's aspect ratio rather than a fixed frame. Don't build a delivery spec around an exact resolution without checking the actual output dimensions first.
  • It adds pixels, not information. Compression blocks, blur and noise already in the source get enlarged along with everything else; very small details may come back reconstructed rather than genuinely recovered.
  • One clip per request, and clips must already be reachable by direct link. A private or expiring URL that the service can't fetch fails before the upscale starts.
  • Price scales with both length and tier. A 60-second clip at 4K costs $1.44, against $0.36 for the same clip at 720p or 1080p — worth picking a smaller tier when the destination doesn't call for more.
  • It can't fix motion problems. Judder, warping or an inconsistent subject from the source generation carries straight through; this only changes resolution, not the underlying footage.

What the results look like

The model page's published example is a 5-second, 832×480 clip of a man climbing a lighthouse staircase by lantern light, upscaled at the default 1080p tier. It returns at 1896×1094 — noticeably larger than the 832×480 source, though short of an exact 1920×1080 frame — with the climb, the lantern glow and the timing of the clip unchanged. That's the trade this endpoint makes in one example: real pixel growth, on the same footage, at a size that tracks the source's aspect ratio rather than landing on a round number.

The takeaway for judging a source clip before you spend the request: like the image upscaler, this works well on footage that's already reasonably sharp and stable. A clip that's soft, heavily compressed or shaky at the source size will still be soft, compressed or shaky at the larger one.

Alternatives

  • Upscaling a still image instead of a clip? Image Upscaler covers 2K, 4K and 8K for photos at a flat $0.012 per image.
  • Want the extra resolution baked in from the start instead of added afterward? Wan 3.0 and Seedance 2.5 both offer higher native output tiers if you're generating the clip rather than upscaling existing footage.
  • Need a frame reframed to a new aspect ratio rather than enlarged? There's no video equivalent of Image Expander yet; that tool currently covers stills only.

How far Video Upscaler goes

Each level, on each task, based on confirmed results. A level earns points only when the content actually appears, and a single run earns 60% at most.

Freedom points by level and task
LevelUpscaling
MildThe lightest prompts in the test
2/2 delivered16.0 / 20 pts
ModerateWhere filtered models start to refuse
5/5 delivered20.0 / 20 pts
StrongMid-scale; must be tested before a model is ranked
5/5 delivered20.0 / 20 pts
Very strongWhere most models start to hold backNot tested yet
MaximumThe top of the scale, scored in two halves: class A and class BNot tested yet
Task score56.0 / 100.0Tested Oct 5, 2026

Tested Oct 5, 2026 to Oct 5, 2026

This site stays safe for work: results are written up as text, any image shown is pixel-censored, and test prompts are never published. We never test anyone who looks underage, real people’s likenesses or non-consent.

See every test record →See the uncensored leaderboard →

Run your own prompt and see how far it goes

Open the Playground

How Video Upscaler earns its Spicy Index

Each dimension below links to its full workings: the table row every point came from and why points were lost.

Public leaderboards 45% of quality
Not yet tested

No exact match on a tracked public leaderboard yet. How it works

Blind output test 35% of quality
56.5
Arena quality 20% of quality
Not yet tested

Counts once the model has 100+ blind votes and a 95% interval under 200 Elo. How it works

Capability 10.5% weight
Not yet tested

Not scored yet. How it works

Freedom Score 30% weight
56.0

Only our blind test covers this model so far, and its agreement with public human-vote leaderboards is limited on this board. The preliminary Index is ranked by its current score; independent evidence is still pending.

Open the full test report →

What Video Upscaler can do on SpicyAPI

Straight from the live model specification on SpicyAPI. Some creators offer more in their own apps; this is what you can call through our API today.

Capabilities by endpoint
FeatureUpscaling
Resolutions720p · 1080p · 2k · 4k
Native audioNo
Seed controlNo
Negative promptNo
See every parameter and code sample →

How long it really takes

Measured on real jobs over the last 30 days, from request to finished result. Times include our own test runs.

Median time
1.7 min
90th percentile
10.1 min
Wait per second of video
Collecting data
Success rate
Collecting data

A time appears once at least 10 completed jobs are in. Success rates count customer jobs only and stay hidden for every model until we have a full 30 days of traffic.

Models to consider instead of Video Upscaler

What do you think?

Was this review useful?

Log in to vote.

Comments

Comments are checked automatically and by our team. We remove spam, offensive content, personal attacks and anything that names a real person without consent. Report a comment and we review it within one business day.

Tried this model? Share what worked. You need a free account to comment, so we can keep spam out.

Video Upscaler FAQ

Try Video Upscaler yourself

Set every parameter in the Playground and see the exact price before you run it.

More reviews

See the leaderboards →
  • A flowing sculptural ribbon and a brass cinema aperture

    Talking Avatar

    Pick Talking Avatar for a fast, cheap way to make a still portrait speak your own audio; it's not built for full-body movement or long-form video.

    Spicy Index
    68.5

    Tested Sep 27, 2026

    Read the review
  • An ivory microphone beside a glass cinema aperture

    Lip Sync

    Pick Lip Sync when you already have footage and just need a new line coming out of the same mouth; it won't fix a clip that isn't a steady, single-shot take.

    Spicy Index
    65.6

    Tested Sep 27, 2026

    Read the review
  • An ivory clapperboard with a sculptural crimson ribbon

    Video Sound Effects

    Pick Video Sound Effects to add ambience and effects to a silent clip cheaply; it won't write dialogue or a music track, so plan for that separately.

    Spicy Index
    —
    Freedom Score
    52.0

    Tested Sep 27, 2026

    Read the review