Flux3ImageToVideoNode - ComfyUI Built-in Node Documentation

Flux 3 Image to Video animates 1 to 10 images with FLUX 3. Each image becomes a frame of the clip: one image opens it, two morph from the first to the second, and more are spread across it or pinned to times you choose.

Inputs

Parameter	Description	Data Type	Required	Range
`prompt`	How the scene should move and sound; the prompt is interpreted and expanded before generation. Must contain at least one character.	STRING	Yes	Multiline text (default: empty)
`keyframes`	1 to 10 images, in playback order. Minimum 256x256 pixels each. Growable input: connect images as `image_1`, `image_2`, and so on.	IMAGE	Yes	1 to 10 images
`placement`	”spread across the clip” lets FLUX 3 place the images (one opens the clip, two become its start and end); “at times” pins every image to a second you choose.	DYNAMIC_COMBO	Yes	`"spread across the clip"` (default) `"at times"`
`times`	One time in seconds per image, comma-separated and increasing, e.g. “0, 2.5, 5”. Only appears when `placement` is “at times”; one time is required for each keyframe image.	STRING	No	Comma-separated seconds (default: “0”)
`aspect_ratio`	Output aspect ratio. “auto” picks one from the prompt and inputs.	COMBO	Yes	`"auto"` (default) other available aspect ratios
`duration`	Clip length in seconds. “auto” fits the length to the content.	COMBO	Yes	`"auto"` (default) other available durations
`resolution`	Output resolution.	COMBO	Yes	`"720p"` (default) `"1080p"`
`generate_audio`	Generate synchronized audio (ambient, speech, effects). Off produces a video with no audio track.	BOOLEAN	Yes	true / false (default: true)
`safety_tolerance`	Moderation tolerance, 0 is the strictest. Requests that send images or video are capped at 2 whatever you set here.	INT	Yes	0 to 4 (default: 2, advanced setting)
`seed`	Seed to determine if node should re-run; FLUX 3 picks its own seed, so actual results are nondeterministic regardless of this value.	INT	Yes	0 to 4294967295 (default: 42, control-after-generate)

Note: keyframes must contain at least one image; the node raises an error if none are connected. Each keyframe image must be at least 256x256 pixels and its aspect ratio cannot be more extreme than 64:1. When placement is “spread across the clip” and 3 or more keyframes are connected, duration must be set to an explicit value, not “auto”; otherwise the node raises an error. When placement is “at times”, times must provide one time in seconds per image. The times must increase, cannot be negative, and the last time cannot be past the end of the clip (up to 20 seconds when duration is “auto”). Because this node sends images, safety_tolerance is capped at 2 regardless of the value you set.

Outputs

Output Name	Description	Data Type
`video`	The generated video clip built from the keyframe images at the chosen aspect ratio, duration, resolution, and audio setting.	VIDEO

This documentation was AI-generated. If you find any errors or have suggestions for improvement, please feel free to contribute! Edit on GitHub

Source fingerprint (SHA-256): 1c49838dfa13adc2ed70a51094f0dd860df7207970b8dceab6bb273653d7161c

​Inputs

​Outputs

Inputs

Outputs