seedinteger
Random seed for reproducibility. If None, a random seed is chosen.
Wan 2.6 Flash Reference to Video
alibaba/wan/2.6/flash/reference-to-videoAlibaba Wan 2.6 Flash reference-to-video model that places characters from reference images or videos into a new scene, with optional generated audio and multi-shot control.
Try the model
Playground
Input
Use Character1, Character2, etc. to reference subjects from your reference files. Works for people, animals, or objects. For multi-shot prompts: '[0-3s] Shot 1. [3-6s] Shot 2.' Max 1500 characters. Reference order: video_urls first, then image_urls.
Additional Settings More
Customize your input with more control.
PNG, JPEG, WebP, or GIF · 20 MiB maximum each
Reference images for subject consistency (0-5 images). Combined with video_urls, total references cannot exceed 5. Formats: JPEG, JPG, PNG (no alpha), BMP, WEBP. Resolution: 240-5000px. Max 10MB each. Reference order: image_urls continue numbering after video_urls.
Reference videos for subject consistency (0-3 videos). Videos' FPS must be at least 16 FPS. Combined with image_urls, total references cannot exceed 5. Reference order: video_urls are numbered first (Character1, Character2...), then image_urls continue the sequence.
Video resolution tier. R2V Flash only supports 720p and 1080p. Allowed values: 720p, 1080p.
Duration of the generated video in seconds. R2V Flash supports only 5 or 10 seconds. Allowed values: 5, 10.
When true (default), enables intelligent multi-shot segmentation for coherent narrative videos with multiple shots. When false, generates single continuous shot. Only active when enable_prompt_expansion is True.
The aspect ratio of the generated image. Allowed values: 16:9, 9:16, 1:1, 4:3, 3:4.
Random seed for reproducibility. If None, a random seed is chosen.
Whether to generate a video with audio. Set to false for silent video generation. Silent videos are faster and cost half of the audio version price.
ResultExample
What would you like to do next?
Examples
Select an example to load its inputs and result into the playground.
- Open result ↗
Input
{ "prompt": "Dance battle between Character1 and Character2." }
Wan 2.6 Flash Reference to Video
The fast, lower-cost Wan 2.6 reference tier. Cast characters from reference images or videos into a new scene for quick iterations; audio generation and multi-shot cuts are on by default.
Fields
| Field | Type | Default | Notes |
|---|---|---|---|
prompt | string, required | — | Text instruction. |
images | image URL[] | — | Reference images, Image 1, Image 2, … |
videos | video URL[] | — | Reference clips, Video 1, Video 2, … |
resolution | 720p / 1080p | 720p | Sets the per-second rate. |
duration | 5 / 10 | 5 | Output length in seconds. |
multi_shots | boolean | true | Let the model cut between several shots. |
aspect_ratio | 16:9 / 9:16 / 1:1 / 4:3 / 3:4 | 16:9 | Output frame shape. |
seed | integer | random | Fix it to compare prompt changes. |
generate_audio | boolean | true | Generate a synchronized audio track; turning it off halves the rate. |
Billing
With audio you pay $0.10 per output second at 720p and $0.15 at 1080p; silent output costs half. When you send reference videos, 5 seconds of input are added to the billed length.
| Output | 720p | 1080p |
|---|---|---|
| 5s with audio | $0.50 | $0.75 |
| 5s silent | $0.25 | $0.375 |
| 5s with audio + reference video | $1.00 | $1.50 |
| 10s with audio | $1.00 | $1.50 |
Failed runs are not charged.
Prompting
- Refer to references by number and give each a role.
- Keep the action simple for fast drafts, then rerun the winner on the non-Flash tier.
Limits
- Durations are 5 or 10 seconds.
Also on SandBase
Other Wan 2.6 endpoints: alibaba/wan/2.6/text-to-video, alibaba/wan/2.6/image-to-video, alibaba/wan/2.6/reference-to-video, alibaba/wan/2.6/flash/image-to-video, alibaba/wan/2.6/text-to-image, alibaba/wan/2.6/image-to-image, alibaba/wan/2.7/text-to-video (newer).
Start building
Send your first request
Submit this model through the SandBase Unified Run API.
# The quoted heredoc keeps Unicode and shell metacharacters unchanged.
result=$(curl --fail-with-body --silent \
-X POST "https://api.sandbase.ai/v1/run" \
-H "Authorization: Bearer $SANDBASE_API_KEY" \
-H "Content-Type: application/json" \
--data-binary @- <<'SANDBASE_JSON'
{
"model": "alibaba/wan/2.6/flash/reference-to-video",
"images": [
"https://static.sandbase.ai/examples/alibaba/wan/2.6/reference-to-video/flash/input_images_0.png"
],
"prompt": "Dance battle between Character1 and Character2.",
"videos": [
"https://static.sandbase.ai/examples/alibaba/wan/2.6/reference-to-video/flash/input_videos_1.mp4"
],
"duration": 5,
"resolution": "720p",
"multi_shots": true,
"aspect_ratio": "16:9",
"generate_audio": true
}
SANDBASE_JSON
)
run_id=$(printf '%s' "$result" | jq -r .id)
for attempt in $(seq 1 120); do
status=$(printf '%s' "$result" | jq -r .status)
case "$status" in completed|failed|timeout) break ;; esac
sleep 2
result=$(curl --fail-with-body --silent \
-H "Authorization: Bearer $SANDBASE_API_KEY" \
"https://api.sandbase.ai/v1/run/$run_id")
done
status=$(printf '%s' "$result" | jq -r .status)
[ "$status" = completed ] || { echo "Generation ended: $status" >&2; exit 1; }
printf '%s\n' "$result"2. Authentication
Create a SandBase API key and set it in your server environment. All requests use Bearer authentication.
export SANDBASE_API_KEY="YOUR_API_KEY"Keep the key on your server. Do not expose it in browser code or commit it to a repository.
3. Queue
Submit a request
POST to /v1/run with this model ID and its inputs. The request example above includes submission and polling.
Fetch request status and result
Use the returned run ID to poll until the status is completed, failed or timeout. Read output only after completed.
curl "https://api.sandbase.ai/v1/run/$RUN_ID" \
-H "Authorization: Bearer $SANDBASE_API_KEY"4. Files
Hosted files
Pass a publicly accessible file URL in the model’s image, video or audio field. Supported formats and limits are shown beside each Playground input.
Uploading images
Use the upload or replace control on supported image fields in the Playground. Video URL fields accept a hosted driving video; they do not imply video-upload support.
5. Schema
Input
imagesarray
Reference images for subject consistency (0-5 images). Combined with video_urls, total references cannot exceed 5. Formats: JPEG, JPG, PNG (no alpha), BMP, WEBP. Resolution: 240-5000px. Max 10MB each. Reference order: image_urls continue numbering after video_urls.
promptstringrequired
Use Character1, Character2, etc. to reference subjects from your reference files. Works for people, animals, or objects. For multi-shot prompts: '[0-3s] Shot 1. [3-6s] Shot 2.' Max 1500 characters. Reference order: video_urls first, then image_urls.
videosarray
Reference videos for subject consistency (0-3 videos). Videos' FPS must be at least 16 FPS. Combined with image_urls, total references cannot exceed 5. Reference order: video_urls are numbered first (Character1, Character2...), then image_urls continue the sequence.
durationinteger
Duration of the generated video in seconds. R2V Flash supports only 5 or 10 seconds.
Default: 5
Possible values: 5, 10
resolutionstring
Video resolution tier. R2V Flash only supports 720p and 1080p.
Default: "720p"
Possible values: "720p", "1080p"
multi_shotsboolean
When true (default), enables intelligent multi-shot segmentation for coherent narrative videos with multiple shots. When false, generates single continuous shot. Only active when enable_prompt_expansion is True.
Default: true
aspect_ratiostring
The aspect ratio of the generated image.
Default: "16:9"
Possible values: "16:9", "9:16", "1:1", "4:3", "3:4"
generate_audioboolean
Whether to generate a video with audio. Set to false for silent video generation. Silent videos are faster and cost half of the audio version price.
Default: true
{
"model": "alibaba/wan/2.6/flash/reference-to-video",
"images": [
"https://static.sandbase.ai/examples/alibaba/wan/2.6/reference-to-video/flash/input_images_0.png"
],
"prompt": "Dance battle between Character1 and Character2.",
"videos": [
"https://static.sandbase.ai/examples/alibaba/wan/2.6/reference-to-video/flash/input_videos_1.mp4"
],
"duration": 5,
"resolution": "720p",
"multi_shots": true,
"aspect_ratio": "16:9",
"generate_audio": true
}Output
A completed run exposes its result in output. The following is the model’s saved example, not a new generation.
[
{
"poster": "https://static.sandbase.ai/examples/posters/162c99e9ecd6e4569806.jpg",
"source": "Alibaba Model Studio schema mirror",
"source_url": "https://www.alibabacloud.com/help/en/model-studio/",
"url": "https://static.sandbase.ai/examples/06adda09-8e45-4ccd-b799-23b878e72dad/output/e57df3be-b22e-4367-a2a5-e88bc35334bb.mp4"
}
]