Conut.aiConut.ai Interactive app Open app

Video Upscaler

Upscale video to 720p, 1080p, 2K or 4K with FlashVSR. 4–10 credits per second, a 5-second minimum, and source clips up to 10 minutes and 200MB.

Video Upscaler raises the resolution of an existing clip to 720p, 1080p, 2K or 4K while reconstructing detail that compression has smoothed away. It uses FlashVSR and is built for frame-to-frame consistency, so faces, edges and fine texture stay stable instead of shimmering. Pricing is 4–10 credits per second depending on the target resolution, with a 5-second minimum and a 10-minute input limit.

How to use Video Upscaler

Choose the lowest resolution that meets the delivery spec, and start from the cleanest source you have.

  1. Find the best source copy A clean 720p source upscales better than a heavily compressed 1080p one. Avoid another compression pass before uploading.
  2. Set the target resolution 720p suits social and drafts, 1080p covers most web delivery, and 2K or 4K are for large screens and masters.
  3. Keep the original frame rate Do not change cadence before upscaling. The model preserves the source timing and will reproduce judder faithfully.
  4. Segment long clips Process a ten-minute video in parts so the work can run in parallel and a failed segment can be re-run alone.
  5. Review faces and edges in motion Check the output at full speed rather than on a still. Temporal flicker on faces and slow camera moves is the main failure mode.

Settings reference

Video Upscaler has one real decision, the target resolution. FlashVSR reconstructs detail and preserves the source frame rate without changing the content.

SettingValueNotes
VideoRequired, up to 200MB and 10 minutesUse the highest-quality copy available; extra compression gives the model more artefacts to fight.
Target resolution720p / 1080p / 2K / 4K1080p is the default. Rates rise from 4 to 10 credits per input second.
PromptNoneThere is no prompt or content control; the tool only raises resolution and clarity.
Price4 / 6 / 8 / 10 credits per input second720p, 1080p, 2K and 4K respectively. A 30-second clip costs 120 at 720p and 300 at 4K.
Minimum charge5 secondsClips shorter than five seconds are billed as five seconds.
Billing basisInput duration, rounded upThe charge follows the source length, not the upscaled file size.
Frame ratePreservedThe source cadence is kept. Changing it before upload can introduce judder.

What upscaling does

The model takes a low-resolution or heavily compressed clip and rebuilds it at a higher resolution. It is not a simple interpolation: edges are reconstructed, texture is re-synthesised and compression artefacts are reduced. FlashVSR's focus is temporal consistency, which matters most on faces and slow camera moves where frame-by-frame flicker is obvious.

This is a restoration tool, not a restyler. It does not change the content, add objects or reinterpret the scene. If the source is soft because of motion blur or a bad lens, upscaling cannot recover information that was never captured; it can only make the existing image cleaner and larger.

Pricing by target resolution

Billing is per second of input video at the chosen output resolution: 4 credits at 720p, 6 at 1080p, 8 at 2K and 10 at 4K. Duration is rounded up to the next second, and there is a 5-second minimum charge, which only matters for very short clips.

A 30-second clip costs 120 credits at 720p, 180 at 1080p and 300 at 4K. A full 10-minute clip costs 2,400 at 720p and 6,000 at 4K, so choosing the lowest resolution that meets the delivery spec is the main cost lever.

Clip length720p1080p2K4K
5s20304050
30s120180240300
60s240360480600
5 min1,2001,8002,4003,000
10 min2,4003,6004,8006,000

Choosing a target resolution

Match the output to where the clip will be seen. 720p is enough for social feeds, embedded players and draft review. 1080p is the safe default for YouTube, client review and most web delivery. 2K and 4K are for large screens, archival masters and footage that will be cropped or reframed in an edit.

Upscaling beyond the source's useful detail has diminishing returns. A 480p phone clip can benefit from 1080p, but pushing it to 4K mostly adds file size. If the source is already 1080p and the destination is 1080p, upscaling will not help unless the goal is compression-artefact repair.

Getting the best source

Start with the highest-quality copy you have. A clean 720p source upscales better than a heavily compressed 1080p one, because the model has less noise to fight. Avoid re-encoding before upload: every extra compression pass removes detail and adds artefacts that the upscaler may amplify.

Keep the original frame rate. Changing cadence before upscaling can introduce judder that the model will faithfully reproduce. For long videos, process in segments and merge afterwards; this parallelises the work and lets you re-run only the segment that failed rather than the whole clip.

Limits and alternatives

Input is limited to 10 minutes and 200MB. There is no prompt and no resolution-specific guidance to write; the only choice is the target resolution. The 5-second minimum means a 2-second clip is billed as five seconds.

Video Upscaler Pro uses the same workflow at 7–17 credits per second, with better frame consistency for difficult footage. Video Extend lengthens a clip rather than sharpening it, and Add Sound to Video creates a Foley track. For prompt-driven picture changes, use Kling Video Edit or WAN 2.2 Video Edit.

Pro tips for Video Upscaler

Most of the cost is resolution multiplied by length. The cheapest correct answer is usually one tier below the maximum.

Frequently asked questions

How much does Video Upscaler cost?

4 credits per second at 720p, 6 at 1080p, 8 at 2K and 10 at 4K, with a 5-second minimum.

How long can the input video be?

Up to 10 minutes and 200MB.

What resolutions can I upscale to?

720p, 1080p, 2K and 4K.

Is a prompt required?

No. The only setting is the target resolution.

What happens with a clip shorter than 5 seconds?

It is billed as 5 seconds, the minimum charge.

Does upscaling change the frame rate?

No. The source cadence is preserved, which is why keeping the original frame rate before upload matters.

Can it fix blurry footage?

It can reduce compression artefacts and reconstruct edges, but it cannot recover detail lost to motion blur, focus or a low-quality lens.

What is the difference from Video Upscaler Pro?

Pro costs more per second (7–17 credits) and targets better frame consistency and detail on difficult footage. Standard is the cheaper default for clean sources.

Compare with other video tools

Specifications

About Conut.ai

Conut.ai is an AI image, video and audio generation platform with 50+ creative tools. It supports text-to-image, image-to-video, face swap, motion control, relighting, inpainting and video effects through models including Nano Banana Pro, Seedream, GPT Image, Sora 2, Kling, WAN and Seedance. Start free — no credit card required.

Explore more AI tools