Skip to main content
Qwen Cloud home page
Docs
Video gen & edits

Video generation models

Choose a model for text-to-video, image-to-video, character animation, and more.

Text-to-video

Prompt-to-video with synchronized audio → wan2.6-t2v. Multi-shot narrative, 1080P, up to 15s per clip.

Need audio?

Synced narration, sound effects, or music → wan2.6-t2v or wan2.5-t2v-preview. Silent video is fine → older models (wan2.2-t2v-plus) cost less.

Resolution and duration

1080P, up to 15s → wan2.6-t2v. 480P–720P, 5s → wan2.2-t2v-plus / wan2.1-t2v-turbo (cheaper).

Image-to-video

Animate a still image into motion → wan2.6-i2v. Budget-friendly → wan2.6-i2v-flash.

Building long, coherent videos?

Use first-and-last-frame models (wan2.2-kf2v-flash) to chain segments: the last frame of one clip becomes the first frame of the next, creating seamless transitions for storytelling, product demos, or tutorials.

Single image to clip

Standard animation from one image → wan2.6-i2v (audio, 1080P, 2–15s). Fast/cheap → wan2.6-i2v-flash.

Reference-to-video

Consistent characters across scenes → wan2.6-r2v. The model replicates appearance from reference videos/images into new scripts with multi-character support. Budget-friendly → wan2.6-r2v-flash (single or multi-character, fast).

Video editing

Redraw, extend, locally edit, or expand frames of existing video → wan2.1-vace-plus.

Character animation

Animate a person from motion reference

Transfer motion from a reference video onto a person in a still image → wan2.2-animate-move. Background stays unchanged. Professional mode (wan-pro): closer to a real-life shot. Standard mode (wan-std): faster and cheaper.

Swap a person in a video

Replace a person in a video with someone from an image → wan2.2-animate-mix. Same pro/std modes.
ModelUse this when...AudioMax resolutionMax duration
wan2.6-t2vBest quality text-to-video✓720P, 1080P2–15s
wan2.6-i2vAnimate an image, highest quality✓720P, 1080P2–15s
wan2.6-i2v-flashAnimate an image, budget-friendly✓720P, 1080P2–15s
wan2.2-kf2v-flashChain clips for long coherent videos—480P–1080P5s
wan2.6-r2vConsistent characters across scenes✓720P, 1080P2–10s
wan2.6-r2v-flashCharacter consistency, budgetopt720P, 1080P2–10s
wan2.1-vace-plusEdit existing video—720Pup to 5s
wan2.2-animate-moveMotion transfer to still person—720P2–30s
wan2.2-animate-mixFace swap in video—720P2–30s

All models

ModelCapabilityFeaturesOutput
wan2.6-t2vText-to-videoAudio sync, multi-shot narrative720P, 1080P. 2–15s. 30 fps, MP4
wan2.6-i2vImage-to-videoAudio sync, multi-shot narrative720P, 1080P. 2–15s. 30 fps, MP4
wan2.6-i2v-flashImage-to-videoAudio, multi-shot, fast720P, 1080P. 2–15s. 30 fps, MP4
wan2.6-r2vReference-to-videoAudio sync, multi-character, narrative720P, 1080P. 2–10s. 30 fps, MP4
wan2.6-r2v-flashReference-to-videoMulti-character, fast720P, 1080P. 2–10s. 30 fps, MP4
ModelCapabilityFeaturesOutput
wan2.5-t2v-previewText-to-videoAudio sync480P, 720P, 1080P. 5s, 10s. 30 fps, MP4
wan2.5-i2v-previewImage-to-videoAudio sync480P, 720P, 1080P. 5s, 10s. 30 fps, MP4
ModelCapabilityFeaturesOutput
wan2.2-t2v-plusText-to-videoNo audio480P, 1080P. 5s. 30 fps, MP4
wan2.2-i2v-plusImage-to-videoNo audio480P, 1080P. 5s. 30 fps, MP4
wan2.2-i2v-flashImage-to-videoNo audio, 50% faster than 2.1480P, 720P, 1080P. 5s. 30 fps, MP4
wan2.2-kf2v-flashFirst & last framesNo audio480P, 720P, 1080P. 5s. 30 fps, MP4
wan2.2-animate-moveCharacter animationwan-std / wan-pro modes720P. 2s–30s. 15/25 fps. MP4
wan2.2-animate-mixCharacter swapwan-std / wan-pro modes720P. 2s–30s. 15/25 fps. MP4
Previous generation. We recommend Wan 2.6 for new projects.
ModelCapabilityFeaturesOutput
wan2.1-t2v-plusText-to-videoNo audio720P. 5s. 30 fps, MP4
wan2.1-t2v-turboText-to-videoNo audio480P, 720P. 5s. 30 fps, MP4
wan2.1-i2v-plusImage-to-videoNo audio720P. 5s. 30 fps, MP4
wan2.1-i2v-turboImage-to-videoNo audio480P, 720P. 3s–5s. 30 fps, MP4
wan2.1-kf2v-plusFirst & last framesNo audio720P. 5s. 30 fps, MP4
wan2.1-vace-plusVideo editingNo audio720P. Up to 5s. 30 fps, MP4

Learn more

Text-to-video

Generate videos from text prompts.

Image-to-video: first frame

Animate from a single image.

First & last frames

Animate between two frames.

Reference-to-video

Generate videos with character consistency.

Video editing

Edit, extend, and redraw videos.
Video generation models | Qwen Cloud