wowmachine
alibaba/wan-2.2-i2v-fast

wan-2.2-i2v-fast

alibaba/wan-2.2-i2v-fast

Wan 2.2 i2v Fast is the PrunaAI-optimised variant of Alibaba's Wan 2.2 A14B image-to-video — engineered for short turnaround and predictable per-video pricing. Drop in a first frame, optionally a last frame, write a short motion prompt, and ship a 480p or 720p clip for ads, social cutdowns or product loops.

Input

Describe the motion and direction of the generated video.

First-frame image to animate.

add_photo_alternate

Optional last-frame image. When set, the model interpolates between the two key frames.

add_photo_alternate

Output resolution. 720p costs roughly 2× more per video than 480p.

Number of frames in the generated video. 81 gives the best results at the native frame rate.

min: 17max: 81

Frame rate of the output video. Pricing is per video, not per frame.

min: 8max: 30

Interpolate the output to 30 FPS using ffmpeg. Adds a small surcharge but produces smoother motion playback.

Optional LoRA weights for the HIGH transformer. Hugging Face slug or `.safetensors` URL.

Strength of the HIGH transformer LoRA.

min: 0max: 2

Optional LoRA weights for the LOW transformer.

Strength of the LOW transformer LoRA.

min: 0max: 2

Run with speed optimisations. Default on.

Optional random seed for reproducible generations.

min: 0max: 2147483647

Disable the NSFW safety filter on the output. Off by default.

Output

movie

Run the model to see output here.

Examples

Pricing

Tiered by **resolution × interpolation**: | Tier | Credits per video | |---|---| | 480p base | 5 | | 480p + interpolate to 30 fps | 7 | | 720p base | 11 | | 720p + interpolate to 30 fps | 15 | *One ...

Features

## What Wan 2.2 i2v Fast is good at Wan 2.2 i2v Fast is PrunaAI's performance-tuned variant of Alibaba's Wan 2.2 A14B image-to-video model. It accepts a first-frame image plus a motion prompt and ret...

API Examples

curl -X POST \
  /api/v1/model/alibaba/wan-2.2-i2v-fast \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
  "input": {}
}'

Tags