Video
Wan VACE 14B
Wan VACE 14B guides video generation or editing with source video, reference images, frames and masks.
- Video editing
Sign in to create. Review the credit cost in the studio before running.
Catalog preview for Wan VACE 14B. Review your own result before publishing.
A starting point.
Use it for a localized video edit or a reframing task where reference material should guide the scene.
Before you publish.
Inspect mask boundaries through motion and ensure the edit does not alter areas you intended to preserve.
Review the quote before running. A failed generation releases its reserved credits. Download the results you want to keep; stored media is retained for 30 days.
Settings and model questionsCheck the available settings and guidance when preparing your input.
Inputs and settings
Provide prompt and select task for the edit. Add video_url, ref_image_urls or endpoint frames as needed; align masks with the source.
Choose 480p, 580p or 720p. Set num_frames and frames_per_second together for pacing; mask_image_url takes precedence over mask_video_url when both are supplied.
Inputs, all modes and APIModel IDs, current input limits and the API generation workflow.
Supported inputs
These fields come from the current studio form for wan-vace-14b. Other modes may require different inputs.
| Input | Type | Requirement | Options and limits |
|---|---|---|---|
prompt | Text | Required | — |
acceleration | Text | Optional | — |
aspect_ratio | Text | Optional | auto, 16:9, 1:1, 9:16 |
auto_downsample_min_fps | Number | Optional | ≥ 1 · ≤ 60 |
enable_auto_downsample | On/off | Optional | — |
enable_prompt_expansion | On/off | Optional | — |
enable_safety_checker | On/off | Optional | — |
first_frame_url | Text | Optional | — |
frames_per_second | Whole number | Optional | ≥ 5 · ≤ 30 |
guidance_scale | Number | Optional | ≥ 1 · ≤ 10 |
interpolator_model | Text | Optional | — |
last_frame_url | Text | Optional | — |
mask_image_url | Text | Optional | — |
mask_video_url | Text | Optional | — |
negative_prompt | Text | Optional | — |
num_frames | Whole number | Optional | ≥ 17 · ≤ 241 |
num_inference_steps | Whole number | Optional | ≥ 2 · ≤ 50 |
num_interpolated_frames | Whole number | Optional | ≥ 0 · ≤ 5 |
preprocess | On/off | Optional | — |
ref_image_urls | List | Optional | — |
resolution | Text | Optional | 480p, 580p, 720p |
return_frames_zip | On/off | Optional | — |
sampler | Text | Optional | — |
seed | Whole number | Optional | — |
shift | Number | Optional | ≥ 1 · ≤ 15 |
task | Text | Optional | — |
temporal_downsample_factor | Whole number | Optional | ≥ 0 · ≤ 5 |
transparency_mode | Text | Optional | — |
video_quality | Text | Optional | — |
video_url | Text | Optional | — |
video_write_mode | Text | Optional | — |
Check the full model schema for all options and cross-field rules. Request a quote for your selected settings before generating.
Use Wan VACE 14B through the API.
Start with this model ID for the endpoint featured on this page:
wan-vace-14b
Input requirements belong to this model ID. Check a different mode’s schema separately when switching modes.
Check the current model schema for supported inputs and options. Other tasks may use different model IDs.
From quote to result.
Request a credit quote for your input and settings before submitting a generation. A sample output or the price of another model is not a quote for your request.
Follow the generation workflow, then poll the job for its status and result. The API and studio use the same balance.