LTX-2.5 API workflows
The template library ships with three API example workflows:- Text to Video: Generate videos from text prompts
- Image to Video: Generate videos from a single start image
- FLF2V: Generate a video between a first and last frame image
LTX-2.5 Text to Video
Generate videos from text prompts through the LTX-2.5 API, with synchronized audio.Run in Comfy Cloud
Open in Comfy Cloud
Download Workflow
Download JSON or search “LTX-2.5 T2V” in Template Library
Prompting tips
- Model: Choose
LTX-2.5 (Pro)for the highest quality, orLTX-2.5 (Fast)for faster generation - Duration: Pro supports 2-10 seconds; Fast supports 2-20 seconds. Durations over 10 seconds require a 720p or 1080p resolution and 24/25 FPS
- Resolution and ratio: Choose a resolution preset such as
1280x720(landscape) or720x1280(portrait); Fast adds2560x1440and3840x2160options - FPS: Choose between 24, 25, 48, or 50 FPS on Fast, and 24, 25, or 50 FPS on Pro
- Audio: Audio generation is on by default; disable it in the node’s advanced settings if you only need a silent video
- Prompt: Describe the whole scene in one flowing paragraph, including the shot, action, characters, camera movement, and the audio you want
LTX-2.5 Image to Video
Generate a video from a single start image through the LTX-2.5 API, with synchronized audio.Run in Comfy Cloud
Open in Comfy Cloud
Download Workflow
Download JSON or search “LTX-2.5 I2V” in Template Library
Input Image: space_battle.png
Start frame for the workflow, or use your own image.
Prompting tips
- Model: This workflow ships with
LTX-2.5 (Fast)selected for faster iteration; switch toLTX-2.5 (Pro)for maximum quality - Single image only: The image input accepts exactly one start image
- Prompt: Describe what happens next: the motion, camera movement, and sounds that follow from the start image. Phrasing like “Use the provided start image as the first frame” helps anchor the generation
- Duration and resolution: The Fast tier supports up to 20 seconds and resolutions up to 4K; durations over 10 seconds require 720p/1080p at 24/25 FPS
LTX-2.5 FLF2V
Generate a video between a first and last frame image through the LTX-2.5 API, blending both frames into a single animated sequence.Run in Comfy Cloud
Open in Comfy Cloud
Download Workflow
Download JSON or search “LTX-2.5 FLF2V” in Template Library
First Frame: porsche_car_01.png
First frame of the video.
Last Frame: porsche_car_02.png
Last frame of the video, connected to the
last_frame input of the node.Prompting tips
- Frames: The workflow connects the first image to the
imageinput and the second image to thelast_frameinput; the model generates the motion between them - Keep frames aligned: Use images with the same aspect ratio for a smooth interpolation
- Model: This workflow ships with
LTX-2.5 (Pro)selected; switch toLTX-2.5 (Fast)for faster generation - Prompt: Describe the transition between the frames, and the audio (dialogue, SFX, music) you want