Wan 2.7 R2V

Uses up to 5 references, including images, video, or audio, to lock hero appearance and voice across shots for episodic content with consistent characters.

Cost

from 6 tokens/s

VideoProvider: Alibaba
Run generation
curl -X POST https://api.givon.ai/api/v1/generations \
  -H "Authorization: Bearer $GIVON_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"type":"video","model":"wan-2.7-r2v","input":{"prompt":"cinematic drone shot over a city at night","aspectRatio":"9:16","resolution":"720p","duration":2,"referenceAssetUris":["asset://asset_..."]}}'

Input fields

* required
prompt*prompt

Scene / motion description.

Type
string
Default
Allowed
string
aspectRatioaspectRatio
Type
string
Default
9:16
Allowed
9:16, 16:9, 1:1, 4:3, 3:4
resolutionresolution
Type
string
Default
720p
Allowed
720p, 1080p
durationduration
Type
number
Default
2
Allowed
2, 3, 4, 5, 6, 7, 8, 9, 10
Start imagestartImage

Asset input for the startImage slot.

Type
string
Default
Allowed
image · asset, https, data
Reference videoreferenceVideo

Asset input for the referenceVideo slot.

Type
string
Default
Allowed
video · asset, https, data
Speech audiospeechAudio

Asset input for the speechAudio slot.

Type
string
Default
Allowed
voiceover · asset, https, data
Reference imagesreferenceAssetUris

Optional reference assets as asset:// URIs, HTTPS URLs, or data URIs.

Type
array
Default
Allowed
image · asset, https, data

Cost

from 6 tokens/s
720pdefault6 tokens/s
1080p8 tokens/s

The variant is selected automatically from request fields, so you do not need to send it.

Capabilities

Modestext_to_videoimage_to_videovideo_to_videoreference_to_video
Asset slotsstartImage:imagereferenceVideo:videospeechAudio:voiceoverreferenceAssetUris:image

Run Wan 2.7 R2V

Get an API key and the same request shape will work across every model in the catalog.

Wan 2.7 R2V API - video generation · Givon AI