Seedvr2 Video | AI Video Upscaler API

SeedVR2 Video Upscaler

Upscale videos to 720p, 1080p, 2K, or 4K with a simple upload. This is the most advanced video upscaler in the world, delivering the highest output quality and exceptional frame-to-frame consistency. If you want a cheaper option, try our Standard Video Upscaler or FlashVSR instead. No local setup required. Also, for Sora-2 wartermark removing, try Watermark Remover!

Why it looks great

Temporal consistency: minimizes flicker and ghosting across frames for stable motion.
Detail reconstruction: restores fine textures (hair, fabric, foliage) and sharp edges without over-sharpening.
Artifact cleanup: reduces compression blocks, ringing, and shimmering in challenging shots.
Motion-aware upscaling: preserves fast action and camera pans with fewer motion artifacts.
Natural look: balances perceptual quality with crispness to avoid plastic or overprocessed outputs.

Limits and Performance

Max clip length per job: up to 10 minutes
Processing speed: approximately 10–30 seconds of wall time to process 1 second of video (varies by resolution and queue load)

Pricing

Per-second billing with a 5-second minimum. The table below lists prices per 5 seconds for easy comparison.

Output Resolution	Cost per 5 seconds
720p	$0.10
1080p	$0.15
2K	$0.20
4K	$0.25

Billing Rules

Minimum charge: 5 seconds
Per-second rate = (price per 5 seconds) ÷ 5
Billed duration = video length in seconds (rounded up), with a 5-second minimum
Total cost = billed duration × per-second rate (by output resolution)

Examples

3.2s @ 1080p → billed as 5s minimum → 5 × $0.03 = $0.15
12s @ 1080p → 12 × $0.03 = $0.36
23s @ 2K → per-second $0.04 → 23 × $0.04 = $0.92
2m01s (121s) @ 4K → per-second $0.05 → 121 × $0.05 = $6.05

How to Use

Choose the target resolution and parameters.
Upload your video (≤ 10 minutes).
Submit the job and wait for processing.
Preview and download the result.

Pro tips for best quality

Upload the highest-quality source you have; avoid heavily compressed inputs when possible.
Keep original frame rate; avoid unnecessary re-encoding before upload.
Pick the lowest resolution that meets your delivery needs (1080p = speed/cost, 2K/4K = maximum detail).
For long videos, process in segments to parallelize and then merge.

Notes

Actual processing time may vary based on resolution, model choice, and current queue.
For videos longer than 10 minutes, split into multiple segments, process separately, and merge afterward.

Seedvr2 Video API — Quick start

Grab a WaveSpeedAI API key, then call POST https://api.wavespeed.ai/api/v3/wavespeed-ai/seedvr2/video with your input as JSON. The endpoint returns a prediction id; poll the prediction endpoint until status flips to completed, then read the output URL from data.outputs[0]. Examples for Seedvr2 Video below.

HTTP example

# Submit the prediction
curl -X POST "https://api.wavespeed.ai/api/v3/wavespeed-ai/seedvr2/video" \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer $WAVESPEED_API_KEY" \
  -d '{
    "video": "https://example.com/your-input.mp4",
    "target_resolution": "1080p"
}'

# Response includes a prediction id. Poll for the result:
curl -X GET "https://api.wavespeed.ai/api/v3/predictions/{request_id}/result" \
  -H "Authorization: Bearer $WAVESPEED_API_KEY"

# When status is "completed", read the output from data.outputs[0].

Node.js example

// npm install wavespeed
const WaveSpeed = require('wavespeed');

const client = new WaveSpeed(); // reads WAVESPEED_API_KEY from env

const result = await client.run("wavespeed-ai/seedvr2/video", {
        "video": "https://example.com/your-input.mp4",
        "target_resolution": "1080p"
});

console.log(result.outputs[0]); // → URL of the generated output

Python example

# pip install wavespeed
import wavespeed

output = wavespeed.run(
    "wavespeed-ai/seedvr2/video",
    {
    "video": "https://example.com/your-input.mp4",
    "target_resolution": "1080p"
}
)

print(output["outputs"][0])  # → URL of the generated output

Seedvr2 Video API — Frequently asked questions

What is the Seedvr2 Video API?

Seedvr2 Video is a WaveSpeedAI model for upscaling, exposed as a REST API on WaveSpeedAI. SeedVR2 Video Upscaler turns low-resolution videos into high-fidelity 4K with seamless motion and consistent frames. Ready-to-use REST inference API, best performance, no coldstarts, affordable pricing. You can call it programmatically or try it from the playground above.

How do I call the Seedvr2 Video API?

POST your input parameters to the model's REST endpoint (shown in the API tab of this playground) with your WaveSpeedAI API key in the Authorization header. Submission returns a prediction ID; poll the prediction endpoint until status flips to "completed", then read the output URL from the result. The playground generates a ready-to-paste code sample in Python, JavaScript, or cURL for whatever inputs you've set. Full request/response shape is documented at https://wavespeed.ai/docs/docs-api/wavespeed-ai/seedvr2-video.

How much does Seedvr2 Video cost per run?

Seedvr2 Video starts at $0.10 per run. That figure is the base price — the final charge scales with the parameters you set in the form (output size, length, count, references, or whatever knobs this model exposes), so a higher-quality or larger output costs more than a minimal one. The exact cost for your current input is shown live next to the Generate button before you submit, and the actual per-call charge is recorded on the prediction afterwards.

What inputs does Seedvr2 Video accept?

Key inputs: `video`, `target_resolution`. The full JSON schema (types, defaults, allowed values) is rendered above the Generate button and mirrored in the API reference at https://wavespeed.ai/docs/docs-api/wavespeed-ai/seedvr2-video.

How long does Seedvr2 Video take to generate?

Average end-to-end generation time on WaveSpeedAI is around 13 seconds per request — measured across recent runs. Queue time scales with global demand; live status is visible in the prediction record.

Can I use Seedvr2 Video outputs commercially?

Commercial usage rights depend on the model's license, set by its provider (WaveSpeedAI). The license summary appears on the model card above; see WaveSpeedAI's Terms of Service for platform-level conditions.

ExamplesView all

Related Models

README