Runway

Runway Gen4 Aleph

Model ID: model_runway-gen4-aleph

Input	Label	Type	Default	Allowed Values	Notes
`prompt`	Prompt	`string`	–	–	required
`video`	Video	`assetId`	–	–	required (<16MB, uses first 5s)
`aspectRatio`	Aspect Ratio	`string`	16:9	9:16, 3:4, 1:1, 4:3, 16:9, 21:9
`referenceImages`	Reference Images	`assetId_array`	–	–	up to 3
`seed`	Seed	`number`	–	–	reproducibility

Pixverse

Pixverse Swap

Model ID: model_pixverse-swap

Input	Label	Type	Default	Min	Max	Allowed Values	Notes
`video`	Video	`assetId`	–	–	–	–	Required. Video to swap person, objects, or background from
`image`	Image	`assetId`	–	–	–	–	Required. Image to swap the subject/background to
`mode`	Mode	`string`	–	–	–	person, background	Selects the swap mode
`resolution`	Resolution	`string`	720p	–	–	720p, 540p, 360p	Output video resolution
`originalSoundSwitch`	Original Sound	`boolean`	true	–	–	–	Includes original sound from input video

LTX

LTX-2 19b Video to Video

Model ID: model_ltx-2-19b-video-to-video

Input	Label	Type	Default	Min	Max	Allowed Values	Notes
`prompt`	Prompt	`string`	None	None	5000	None	Required. Text prompt for video generation.
`video`	Input Video	`file`	None	None	None	None	Required. Reference video. The model generates output conditioned on this video's structure.
`videoStrength`	Video Strength	`number`	1	0	1	None	Reference video conditioning strength (0-1). Higher = more faithful to input video structure.
`image`	First Frame Image	`file`	None	None	None	None	Optional first-frame image. Applied at frame 0 for additional conditioning.
`imageStrength`	Image Strength	`number`	1	0	1	None	First-frame image conditioning strength (0-1).
`negativePrompt`	Negative Prompt	`string`	None	None	5000	None	Negative prompt to guide what to avoid in generation.
`resolution`	Resolution	`string`	auto	None	None	auto, 720p, 1080p, 1440p, 2160p	Video resolution. 'Match Input' replicates input video dimensions (recommended).
`width`	Width	`number`	None	64	4096	None	Explicit video width in pixels. Must be used with height. Rounded to multiple of 64.
`height`	Height	`number`	None	64	4096	None	Explicit video height in pixels. Must be used with width. Rounded to multiple of 64.
`cameraMotion`	Type	`string`	none	None	None	none, static, dolly_in, dolly_out, dolly_left, dolly_right, jib_up, jib_down	Camera movement type in the generated video.
`cameraMotionStrength`	Strength	`number`	1	0	2	None	Strength of camera motion (0-2). Higher values = stronger effect.
`duration`	Duration	`number`	None	3	20	None	Output video duration in seconds (3-20). Leave empty to match input video.
`aspectRatio`	Aspect Ratio	`string`	auto	None	None	auto, 21:9, 16:9, 4:3, 1:1, 3:4, 9:16	Video aspect ratio. 'Match Input' detects from input video (recommended).
`fps`	Frame Rate	`number`	None	1	60	None	Output frame rate (1-60 fps). Leave empty to match input video.
`audioSource`	Audio Source	`string`	original	None	None	original, generate, none	Audio for output: copy from input, generate new, or none.
`modality`	Modality Type	`string`	none	None	None	none, canny, depth, pose, detailer	IC-LoRA control: canny (edges), depth, pose (skeleton), detailer, or none.
`modalityStrength`	Modality Strength	`number`	1	0	2	None	Strength of IC-LoRA modality (0-2). Higher values = stronger control.
`conditioningVideo`	Conditioning Video	`file`	None	None	None	None	Pre-processed conditioning video (canny/depth/pose). Use with matching modality.
`numInferenceSteps`	Steps	`number`	40	1	100	None	Number of inference steps. Higher = better quality, slower.
`guidanceScale`	Guidance	`number`	3	1	20	None	Classifier-free guidance scale. Higher = stronger prompt adherence.
`enhancePrompt`	Enhance Prompt	`boolean`	false	None	None	None	Use Gemma to enhance the prompt for better video generation.
`seed`	Seed	`number`	None	0	None	None	Random seed for reproducibility. Leave blank for random.

Luma

Luma Modify Video

Model ID: model_luma-modify-video

Input	Label	Type	Default	Allowed Values	Notes
`video`	Video	`assetId`	–	–	required (<100MB, <30s)
`firstFrame`	First Frame	`assetId`	–	–	optional reference frame
`prompt`	Prompt	`string`	–	–	–	guides modification
`mode`	Mode	`string`	adhere_1	adhere_1, adhere_2, adhere_3, flex_1, flex_2, flex_3, reimagine_1, reimagine_2, reimagine_3	controls degree of change

Luma Video Reframe

Model ID: model_luma-reframe-video

Input	Label	Type	Default	Min	Max	Allowed Values	Notes
`video`	Video	`assetId`	None	None	None	None	Required. Video to reframe. Maximum duration is 10 seconds
`aspectRatio`	Output Aspect Ratio	`string`	16:9	None	None	9:21, 9:16, 3:4, 1:1, 4:3, 16:9, 21:9	Aspect ratio of the output video
`prompt`	Prompt	`string`	None	None	2048	None	Optional prompt to guide the reframing generation

Wan

Wan 2.2 Animate Replace

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

Input	Label	Type	Default	Min	Max	Allowed Values	Notes
`videoUrl`	Video	`assetId`	–	–	–	–	Required input video
`imageUrl`	Image	`assetId`	–	–	–	–	Required input image. If aspect ratio mismatches, it is resized and center cropped
`resolution`	Resolution	`string`	720p	–	–	720p, 580p, 480p	Output resolution
`numInferenceSteps`	Steps	`number`	30	2	40	–	Number of inference steps
`guidanceScale`	Guidance	`number`	5	1	10	–	Guidance scale for generation
`videoQuality`	Video Quality	`string`	high	–	–	maximum, high, medium, low	Output video quality
`videoWriteMode`	Video Write Mode	`string`	balanced	–	–	balanced, fast, small	Video writing mode
`seed`	Seed	`number`	–	–	–	–	Optional random seed for reproducibility

Wan 2.2 Animate Move

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

Input	Label	Type	Default	Min	Max	Allowed Values	Notes
`videoUrl`	Video	`assetId`	–	–	–	–	Required input video
`imageUrl`	Image	`assetId`	–	–	–	–	Required input image. If aspect ratio mismatches, it is resized and center cropped
`resolution`	Resolution	`string`	720p	–	–	720p, 580p, 480p	Output resolution
`numInferenceSteps`	Steps	`number`	30	2	40	–	Number of inference steps
`guidanceScale`	Guidance	`number`	5	1	10	–	Guidance scale for generation
`videoQuality`	Video Quality	`string`	high	–	–	maximum, high, medium, low	Output video quality
`videoWriteMode`	Video Write Mode	`string`	balanced	–	–	balanced, fast, small	Video writing mode
`seed`	Seed	`number`	–	–	–	–	Optional random seed for reproducibility

Video Foreground Extractor

BiRefNet v2

Model ID: model_birefnet-v2-video

Parameter	Type	Default	Description
`video`	`assetId`	``	Source video for foreground extraction.
`model`	`string`	`General Use (Light)`	Model to use for background removal. The 'General Use (Light)' model is the original model used in the BiRefNet repository. The 'General Use (Light)' model is the original model used in the BiRefNet repository but trained with 2K images. The 'General Use (Heavy)' model is a slower but more accurate model. The 'Matting' model is a model trained specifically for matting images. The 'Portrait' model is a model trained specifically for portrait images. The 'General Use (Light)' model is recommended for most use cases. `"allowedValues": ["General Use (Light)", "General Use (Light 2K)", "General Use (Heavy)", "Matting", "Portrait"]`
`operatingResolution`	`string`	`1024x1024`	The resolution to operate on. The higher the resolution, the more accurate the output will be for high res input images. `"allowedValues": ["1024x1024", "2048x2048"]`
`outputMask`	`boolean`		Whether to output the mask used to remove the background
`refineForeground`	`boolean`		Refine the foreground for better results
`videoQuality`	`string`	`high`	Video output quality. `"allowedValues": ["low", "medium", "high", "maximum"]`
`videoWriteMode`	`string`	`balanced`	Video write mode for encoding. `"allowedValues": ["fast", "balanced", "small"]`

Decart

Lucy Edit Pro

Model ID: model_lucy-edit-pro

Input	Label	Type	Default	Allowed Values	Notes
`prompt`	Prompt	`string`	–	–	Text description of the desired video content - required
`videoUrl`	Video	`assetId`	–	–	Input video to edit
`resolution`	Resolution	`string`	-	480p, 720p
`enhancePrompt`	Enhance Prompt	`boolean`	true	true, false	Whether to enhance the prompt for better results

Lucy Edit Dev

Model ID: model_lucy-edit-dev

Input	Label	Type	Default	Allowed Values	Notes
`prompt`	Prompt	`string`	–	–	Text description of the desired video content - required
`videoUrl`	Video	`assetId`	–	–	Input video to edit
`enhancePrompt`	Enhance Prompt	`boolean`	true	true, false	Whether to enhance the prompt for better results