Alibaba

This page is auto-generated from model configurations. Last updated: 2026-03-13.

This reference lists all available Alibaba video generation models and their parameters. Use these parameter names when calling the Generation API.


Wan 2.1 - 1.3b

Wan 2.1 1.3b is a text-to-video model that ensures detail and accuracy in animations.

Model ID: model_wan-2-1-1-3b

Capabilities: txt2video

LLM Markdown: https://app.scenario.com/api/models/model_wan-2-1-1-3b/markdown

ParameterTypeRequiredDefaultMinMaxAllowed ValuesDescription
promptstringYes----Describe your video
aspectRatiostringNo16:9--16:9, 9:16Video aspect ratio
frameNumnumberNo81--17, 33, 49, 65, 81Video duration in frames (based on standard 16fps playback)
sampleStepsnumberNo301050-Number of sampling steps (higher = better quality but slower)
sampleGuideScalenumberNo6020-Higher values follow the prompt more closely, lower values are more creative
sampleShiftnumberNo8020-Sampling shift factor for flow matching (recommended range: 8-12)
seednumberNo----Use a seed for reproducible results. Leave blank to use a random seed.

Wan 2.2 - I2V

Wan 2.2 A14B is a image-to-video model at 720p and 480p resolutions

Model ID: model_wan-2-2-i2v-a14b

Capabilities: img2video

LLM Markdown: https://app.scenario.com/api/models/model_wan-2-2-i2v-a14b/markdown

ParameterTypeRequiredDefaultMinMaxAllowed ValuesDescription
promptstringYes----Describe your video
imagefileNo----Image used as the first frame of the video. Ideal images are 16:9 or 9:16 and 1280x720 or 720x1280, depending on the aspect ratio you choose.
lastFrameImagefileNo----Input image for last frame generation. This only works if an image start frame is given too.
resolutionstringNo720p--720p, 480pVideo resolution
numFramesnumberNo8181100-Number of video frames. 81 frames give the best results
framesPerSecondnumberNo16524-Frames per second.
sampleStepsnumberNo30150-Number of generation steps. Fewer steps means faster generation, at the expensive of output quality. 30 steps is sufficient for most prompts
sampleShiftnumberNo5120-Controls how much motion is added between video frames. Higher values create faster motion, lower values result in smoother, slower changes.
seednumberNo----Use a seed for reproducible results. Leave blank to use a random seed.

Wan 2.2 - T2V

Wan 2.2 A14B text-to-video

Model ID: model_wan-2-2-t2v

Capabilities: txt2video

LLM Markdown: https://app.scenario.com/api/models/model_wan-2-2-t2v/markdown

ParameterTypeRequiredDefaultMinMaxAllowed ValuesDescription
promptstringYes----Describe your video
resolutionstringNo720p--720p, 480pVideo resolution
numFramesnumberNo8181121-Number of video frames. 81 frames give the best results
framesPerSecondnumberNo16530-Frames per second.
sampleShiftnumberNo5120-Controls how much motion is added between video frames. Higher values create faster motion, lower values result in smoother, slower changes.
seednumberNo----Use a seed for reproducible results. Leave blank to use a random seed.

Wan 2.2 Animate - Move

Wan-Animate is a video model that generates high-fidelity character videos by replicating the expressions and movements of characters from reference videos.

Model ID: model_wan-2-2-14b-animate-move

Capabilities: video2video

LLM Markdown: https://app.scenario.com/api/models/model_wan-2-2-14b-animate-move/markdown

ParameterTypeRequiredDefaultMinMaxAllowed ValuesDescription
videoUrlfileYes----Input video
imageUrlfileYes----Input image. If the input image does not match the chosen aspect ratio, it is resized and center cropped.
resolutionstringNo720p--720p, 480pOutput video resolution
mergeAudiobooleanNotrue---Merge audio from input video into output
numInferenceStepsnumberNo12140-Number of inference steps. Higher values improve quality but slow generation
guidanceScalenumberNo1120-Guidance scale for generation
seednumberNo----Random seed for reproducibility. If None, a random seed is chosen.

Wan 2.2 Animate - Replace

Wan-Animate Replace is a model that can integrate animated characters into reference videos, replacing the original character while preserving the scene's lighting and color tone for seamless environmental integration.

Model ID: model_wan-2-2-14b-animate-replace

Capabilities: video2video

LLM Markdown: https://app.scenario.com/api/models/model_wan-2-2-14b-animate-replace/markdown

ParameterTypeRequiredDefaultMinMaxAllowed ValuesDescription
videoUrlfileYes----Input video
imageUrlfileYes----Input image. If the input image does not match the chosen aspect ratio, it is resized and center cropped.
resolutionstringNo720p--720p, 480pOutput video resolution
mergeAudiobooleanNotrue---Merge audio from input video into output
numInferenceStepsnumberNo12140-Number of inference steps. Higher values improve quality but slow generation
guidanceScalenumberNo1120-Guidance scale for generation
seednumberNo----Random seed for reproducibility. If None, a random seed is chosen.

Wan 2.2 Outpainting

VACE Fun for Wan 2.2 A14B from Alibaba-PAI

Model ID: model_wan-2-2-vace-fun-a14b-outpainting

Capabilities: video2video

LLM Markdown: https://app.scenario.com/api/models/model_wan-2-2-vace-fun-a14b-outpainting/markdown

ParameterTypeRequiredDefaultMinMaxAllowed ValuesDescription
promptstringYes----Text prompt for video generation
videoUrlfileYes----Input video for outpainting
expandLeftbooleanNotrue---Expand video to the left
expandRightbooleanNotrue---Expand video to the right
expandTopbooleanNotrue---Expand video to the top
expandBottombooleanNotrue---Expand video to the bottom
expandRationumberNo0.2501-Amount of expansion. This is a float value between 0 and 1, where 0.25 adds 25% to the original video size on the specified sides.
negativePromptstringNo----Text negative prompt for video generation
refImageUrlsfile_arrayNo``---Reference images
matchInputNumFramesbooleanNofalse---Match the number of frames from input video
numFramesnumberNo8181241-Number of frames to generate
matchInputFramesPerSecondbooleanNofalse---If true, the frames per second of the generated video will match the input video. If false, the frames per second will be determined by the Frames Per Seconds parameter.
framesPerSecondnumberNo16530-Frames per second of the generated video. Ignored if match_input_frames_per_second is true. Default value: 16
resolutionstringNo720p--720p, 580p, 480pOutput video resolution
aspectRatiostringNoauto--auto, 16:9, 1:1, 9:16Aspect ratio for output video
numInferenceStepsnumberNo30250-Number of inference steps
guidanceScalenumberNo5110-Guidance scale for generation
enablePromptExpansionbooleanNofalse---Enable prompt expansion
accelerationstringNoregular--regular, noneProcessing acceleration mode
videoQualitystringNohigh--maximum, high, medium, lowOutput video quality
videoWriteModestringNobalanced--balanced, fast, smallVideo writing mode
numInterpolatedFramesnumberNo105-Number of frames to interpolate between the original frames. A value of 0 means no interpolation
interpolatorModelstringNofilm--film, rifeInterpolator model to use
seednumberNo----Random seed for reproducibility. If None, a random seed is chosen.

Wan 2.2 Reframe

VACE Fun for Wan 2.2 A14B from Alibaba-PAI

Model ID: model_wan-2-2-vace-fun-a14b-reframe

Capabilities: video2video

LLM Markdown: https://app.scenario.com/api/models/model_wan-2-2-vace-fun-a14b-reframe/markdown

ParameterTypeRequiredDefaultMinMaxAllowed ValuesDescription
videoUrlfileYes----Input video for reframe
promptstringNo----Text prompt for video generation
negativePromptstringNo----Text negative prompt for video generation
matchInputNumFramesbooleanNofalse---Match the number of frames from input video
numFramesnumberNo8181241-Number of frames to generate
matchInputFramesPerSecondbooleanNofalse---If true, the frames per second of the generated video will match the input video. If false, the frames per second will be determined by the Frames Per Seconds parameter.
framesPerSecondnumberNo16530-Frames per second of the generated video. Ignored if match_input_frames_per_second is true. Default value: 16
resolutionstringNo720p--720p, 580p, 480pOutput video resolution
aspectRatiostringNoauto--auto, 16:9, 1:1, 9:16Aspect ratio for output video
numInferenceStepsnumberNo30250-Number of inference steps
guidanceScalenumberNo5110-Guidance scale for generation
enablePromptExpansionbooleanNofalse---Enable prompt expansion
accelerationstringNoregular--regular, noneProcessing acceleration mode
videoQualitystringNohigh--maximum, high, medium, lowOutput video quality
videoWriteModestringNobalanced--balanced, fast, smallVideo writing mode
numInterpolatedFramesnumberNo105-Number of frames to interpolate between the original frames. A value of 0 means no interpolation
interpolatorModelstringNofilm--film, rifeInterpolator model to use
zoomFactornumberNo000.9-Zoom factor for the video. When this value is greater than 0, the video will be zoomed in by this factor (in relation to the canvas size,) cutting off the edges of the video. A value of 0 means no zoom.
trimBordersbooleanNotrue---Whether to trim borders from the video.
temporalDownsampleFactornumberNo005-Temporal downsample factor for the video. This is an integer value that determines how many frames to skip in the video. A value of 0 means no downsampling. For each downsample factor, one upsample factor will automatically be applied.
seednumberNo----Random seed for reproducibility. If None, a random seed is chosen.

Wan 2.5 - I2V

Wan 2.5 image-to-video model.

Model ID: model_wan-2-5-i2v

Capabilities: img2video

LLM Markdown: https://app.scenario.com/api/models/model_wan-2-5-i2v/markdown

ParameterTypeRequiredDefaultMinMaxAllowed ValuesDescription
imagefileYes----Image to use for your video
promptstringYes----A textual prompt to guide model generation.
audiofileNo----Audio file for voice/music synchronization. 3-30s, ≤15MB.
negativePromptstringNo----Negative prompt used to guide the model away from undesirable features.
resolutionstringNo720p--720p, 1080pVideo resolution.
durationnumberNo5--5, 10Duration of the generated video in seconds.
enablePromptExpansionbooleanNotrue---Whether to enable prompt rewriting using LLM.
seednumberNo----Random seed for reproducibility. If None, a random seed is chosen.

Wan 2.5 - T2V

Wan 2.5 text-to-video model.

Model ID: model_wan-2-5-t2v

Capabilities: txt2video

LLM Markdown: https://app.scenario.com/api/models/model_wan-2-5-t2v/markdown

ParameterTypeRequiredDefaultMinMaxAllowed ValuesDescription
promptstringYes----A textual prompt to guide model generation.
audiofileNo----Audio file for voice/music synchronization. 3-30s, ≤15MB.
negativePromptstringNo----Negative prompt used to guide the model away from undesirable features.
sizestringNo1280*720--1280*720, 720*1280, 1920*1080, 1080*1920Video resolution and aspect ratio.
durationnumberNo5--5, 10Duration of the generated video in seconds.
enablePromptExpansionbooleanNotrue---Whether to enable prompt rewriting using LLM.
seednumberNo----Random seed for reproducibility. If None, a random seed is chosen.

Wan 2.6 I2V

Alibaba Wan 2.6 image to video generation model

Model ID: model_wan-2-6-i2v

Capabilities: img2video

LLM Markdown: https://app.scenario.com/api/models/model_wan-2-6-i2v/markdown

ParameterTypeRequiredDefaultMinMaxAllowed ValuesDescription
imagefileYes----Input image for video generation
promptstringYes----Text prompt for video generation
audiofileNo----Audio file (3-30s, ≤15MB) for voice/music synchronization
negativePromptstringNo----Negative prompt to avoid certain elements
resolutionstringNo720p--720p, 1080pVideo resolution
durationnumberNo5--5, 10, 15Duration of the generated video in seconds
enablePromptExpansionbooleanNotrue---If set to true, the prompt optimizer will be enabled
multiShotsbooleanNotrue---Enable intelligent multi-shot segmentation (only active when 'Enable Prompt Expansion' is enabled). True enables multi-shot segmentation, false generates single-shot content.
seednumberNo----Random seed for reproducible generation

Wan 2.6 T2V

Alibaba Wan 2.6 text to video generation model

Model ID: model_wan-2-6-t2v

Capabilities: txt2video

LLM Markdown: https://app.scenario.com/api/models/model_wan-2-6-t2v/markdown

ParameterTypeRequiredDefaultMinMaxAllowed ValuesDescription
promptstringYes----Text prompt for video generation
audiofileNo----Audio file (3-30s, ≤15MB) for voice/music synchronization
negativePromptstringNo----Negative prompt to avoid certain elements
sizestringNo1280*720--1280*720, 720*1280, 1920*1080, 1080*1920Video resolution and aspect ratio
durationnumberNo5--5, 10, 15Duration of the generated video in seconds
enablePromptExpansionbooleanNotrue---If set to true, the prompt optimizer will be enabled
multiShotsbooleanNotrue---Enable intelligent multi-shot segmentation (only active when 'Enable Prompt Expansion' is enabled). True enables multi-shot segmentation, false generates single-shot content.
seednumberNo----Random seed for reproducible generation