First-Last Frame To Video

Create video via frames interpolation

Transform two images into a seamless video sequence with the LTX 2.3 Distilled model. This powerful First-Last Frame to Video (FLF2V) technology creates dynamic motion by intelligently interpolating the visual information between a selected starting frame and ending frame. The model analyzes both input images and generates a smooth, coherent transition, bringing static images to life with natural movement, cinematic flow, and consistent visual details. Simply provide the first and last frames, and let LTX 2.3 Distilled FLF2F generate a complete video journey between them.

First image

Second image

Width (px) 704

Height (px) 704

FPS 25

Length (frames) 500

Seamless Motion Generation Between Two Visual Anchors

Generate structured AI video sequences with a duration of up to 24 seconds

LTX 2.3 Distilled FLF2F introduces a new approach to image-to-video creation by using two selected visual anchors: the first frame that defines the beginning of the story and the last frame that defines the final moment. The model predicts and constructs the motion path between them. This interpolation process creates smooth transitions, realistic movement, and a continuous visual narrative while preserving the key elements, style, and identity of the original images.

Creative Control Through First and Last Frame Conditioning

Transform Static Images into Controlled Cinematic Sequences

With First-Last Frame to Video technology, creators gain precise control over the direction of generated videos. By defining the initial and final appearance, LTX 2.3 Distilled FLF2F can transform simple image pairs into cinematic sequences with natural progression and coherent motion. Whether creating visual effects, product animations, concept scenes, or artistic transformations, the model provides a flexible way to guide video generation while maintaining consistency throughout the entire sequence.