Wan 3.0 Prime

Description

Faster Wan 3.0 variant — same multimodal stack, accelerated generation.

Wan. Video generation from Alibaba

Capabilities

Vendor

Alibaba

Family

Wan

Type

Video

Model ID

Tasks

fal.ai: Text to video, Image to video, Reference to video · Pika: Text to video, Image to video, Reference to video · Venice: Text to video, Image to video, Reference to video · WaveSpeed: Text to video, Image to video, Reference to video, Video to video

Seed

fal.ai: Can be set. · Pika: Can be set. · WaveSpeed: Can be set.

Safety checker

fal.ai: Switch.

Extend

WaveSpeed: forward

Video edit

WaveSpeed: Prompt edit.

Inputs

Prompt

Optional

Image

Optional

End frame

Supported

Max images

10

Reference images

Up to 10 reference images.

Reference videos

Up to 5 reference videos.

Reference audio

Up to 5 reference audio clips.

Reference names

Prompts name references in words, not with @ tokens.

Reference required

Yes.

Keyframes

Cannot be combined with references.

Reference video length

Up to 15 seconds.

Reference audio length

Up to 15 seconds.

Video

Optional

Source video

Accepted.

Outputs

Duration

fal.ai: 2–30 seconds · Pika: 2–30 seconds · Venice: 2 seconds, 5 seconds, 10 seconds, 15 seconds, 20 seconds, 25 seconds, 30 seconds · WaveSpeed: 2–30 seconds

Resolution

fal.ai: 480p, 720p, 1080p · Pika: 480p, 720p, 1080p · Venice: 480p, 720p, 1080p · WaveSpeed: 480p, 720p, 1080p

Aspect ratios

fal.ai: adaptive, 16:9, 4:3, 1:1, 3:4, 9:16 · Pika: adaptive, 16:9, 4:3, 1:1, 3:4, 9:16 · Venice: 16:9, 9:16, 1:1, 4:3, 3:4 · WaveSpeed: adaptive, 16:9, 4:3, 1:1, 3:4, 9:16

Audio

fal.ai: Generated, and the catalog records a switch. · Pika: Generated, and the catalog records a switch. · Venice: Generated, and the catalog records a switch. · WaveSpeed: Generated, and the catalog records a switch.