Video Generation

Generate, transform, upscale, and process videos with AI models.

POST/api/v1/video/generate

Request Body

Parameter	Type	Required	Description
model	string	Yes	Model key (see models below)
prompt	string	No	Text prompt (required for most models)
params	object	No	Model-specific parameters
attachmentMediaIds	string[]	No	Positional media attachment IDs

Response

{
  "inferenceJobId": "ij_..."
}

Example

POST/api/v1/video/generate

{
  "model": "sora2",
  "prompt": "A drone shot flying over a misty forest at sunrise",
  "params": { "aspectRatio": "16:9", "duration": "8" }
}

Models

sora2

Text-to-video generation. Prompt required.

Parameter	Type	Required	Description
aspectRatio	string	No	16:9, 9:16
resolution	string	No	720p only
duration	string	No	"4", "8", "12", "16", or "20" seconds (default "8")
omitFirstScene	boolean	No	Omit the first scene from the output

Attachments: 1 optional image.

sora2pro

Pro-tier text-to-video. Prompt required.

Parameter	Type	Required	Description
aspectRatio	string	No	16:9, 9:16
resolution	string	No	720p, 1024p, 1080p (default 720p)
duration	string	No	"4", "8", "12", "16", or "20" seconds (default "8")
omitFirstScene	boolean	No	Omit the first scene from the output

Attachments: 1 optional image.

veo3_1-fast

Fast video generation with fixed 8-second duration. Prompt required.

Parameter	Type	Required	Description
aspectRatio	string	No	16:9, 9:16
resolution	string	No	720p, 1080p

Duration: Fixed at 8 seconds.

Attachments: Up to 2 images (first_frame, last_frame). last_frame requires first_frame to be provided.

kling2_5-pro

Kling 2.5 Pro video generation. Prompt required.

Parameter	Type	Required	Description
aspectRatio	string	No	16:9, 9:16, 1:1
duration	number	No	5 or 10 seconds

Attachments: Up to 2 optional images.

kling_2_6_pro

Kling 2.6 Pro with automatic image-to-video. Prompt required.

Parameter	Type	Required	Description
aspectRatio	string	No	16:9, 9:16, 1:1
duration	number	No	5 or 10 seconds

Attachments: 1 optional image (auto image-to-video when provided).

kling_2_6_motion_control

Motion-controlled video from a reference image and video. Prompt required.

Parameter	Type	Required	Description
characterOrientation	string	No	image, video
keepOriginalSound	boolean	No	Keep the original audio from the reference video

Attachments: 2 required -- reference_image + reference_video.

kling_3_pro

Text-to-video and image-to-video with fine-grained controls. Prompt required.

Parameter	Type	Required	Description
aspectRatio	string	No	Aspect ratio of the output video
duration	number	No	3 to 15 seconds
generateAudio	boolean	No	Generate audio for the video
multiPrompt	boolean	No	Enable multi-prompt mode
negativePrompt	string	No	Negative prompt to guide generation away from
cfgScale	number	No	Classifier-free guidance scale (0 to 1)

Attachments: Optional start_image, end_image, and up to 2 element slots.

sync-lipsync-v2-pro

Lip-sync a video to an audio track. No prompt needed.

Parameter	Type	Required	Description
syncMode	string	No	cut_off, loop, bounce, silence, remap
videoDurationSeconds	number	No	1 to 600 seconds

Attachments: 2 required -- video + audio.

bria-video-background-removal

Remove background from video. No prompt or params needed.

Params: None.

Attachments: 1 required video.

veed-video-background-removal

VEED-powered video background removal. No prompt needed.

Parameter	Type	Required	Description
subjectIsPerson	boolean	No	Whether the subject is a person (optimizes removal)

Attachments: 1 required video.

veed-video-background-removal-fast

Fast variant of VEED video background removal. No prompt needed.

Parameter	Type	Required	Description
subjectIsPerson	boolean	No	Whether the subject is a person (optimizes removal)

Attachments: 1 required video.

seedance-2

Text-to-video and image-to-video generation. Routes automatically based on image attachment.

Parameter	Type	Required	Description
aspectRatio	string	No	16:9, 9:16, 4:3, 3:4
duration	string	No	"5", "10", or "15" (default "5")
quality	string	No	high (standard), basic (fast)

Attachments: 1 optional image (triggers image-to-video mode).

seedance-2-omni

Reference-based video generation. Use @image1, @audio1 in prompt.

Parameter	Type	Required	Description
aspectRatio	string	No	16:9, 9:16, 1:1, 4:3, 3:4, 21:9
duration	string	No	"4" to "15" seconds (default "5")

Attachments: Up to 3 images and 1 audio clip (all optional).

seedance-2-video-edit

Edit an existing video using text prompts.

Parameter	Type	Required	Description
aspectRatio	string	No	16:9, 9:16, 4:3, 3:4
quality	string	No	basic, high (default basic)

Attachments: 1 required video.

grok-imagine-video

Video generation by xAI. Text-to-video and image-to-video. Prompt required.

Parameter	Type	Required	Description
aspectRatio	string	No	9:16, 16:9, 1:1, 4:3, 3:4, 3:2, 2:3 (default 9:16)
duration	string	No	5, 8, 10, or 15 seconds (default 8)
resolution	string	No	480p, 720p (default 720p)

Attachments: 1 optional start image for image-to-video.

topaz-video-upscale

Upscale video resolution with Topaz. No prompt needed.

Parameter	Type	Required	Description
upscaleFactor	number	No	1 to 4
targetFps	string \| number	No	"original" or 16 to 60
codec	string	No	h265, h264

Attachments: 1 required video.