lightricks/ltx-2.3-pro

High-fidelity video generation with portrait support, audio-to-video, retake, and extend. Text, image, and audio-driven creation up to 4K at 50 FPS.

507 runs
License
Playground API Examples README

LTX 2.3 Pro

LTX 2.3 Pro is the high-fidelity variant of Lightricks’ LTX-2.3 video generation model. It produces higher visual quality than LTX 2.3 Fast and includes additional capabilities: audio-to-video, retake (re-generate sections of a video), and extend (add duration to a video).

Tasks

Use the task input to select what you want to do:

Task Description
text_to_video Generate video from a text prompt
image_to_video Generate video from a start frame image (and optional end frame)
audio_to_video Generate video driven by an audio file (1080p only)
retake Re-generate a section of an existing video
extend Add duration to the beginning or end of a video

What’s new in 2.3

  • Portrait video — native 9:16 aspect ratio alongside 16:9 landscape
  • More frame rates — 24, 25, 48, and 50 FPS
  • End frame interpolation — provide both a start and end frame image for image-to-video
  • Camera motion controls — dolly, jib, static, focus shift, and more
  • Improved quality — better VAE, larger text connector for improved prompt adherence, fewer audio artifacts

Inputs

Common inputs

Input Description Default
task Generation task (see above) text_to_video
prompt Text description of the video (required)
resolution 1080p, 2k, or 4k (audio_to_video/retake/extend are 1080p only) 1080p
duration Length in seconds: 6, 8, or 10 6
aspect_ratio 16:9 or 9:16 16:9
fps Frame rate: 24, 25, 48, or 50 25
camera_motion Camera effect: dolly_in, dolly_out, dolly_left, dolly_right, jib_up, jib_down, static, focus_shift, or none none
generate_audio Generate synchronized audio true

Image-to-video inputs

Input Description
image First frame image (required for image_to_video)
last_frame_image Optional end frame for interpolation

Audio-to-video inputs

Input Description
audio Audio file (.wav, .mp3, .m4a, .ogg)

Retake inputs

Input Description Default
video Input video to edit (required)
retake_start_time Start time in seconds of section to edit 0.0
retake_duration Duration in seconds of section to edit (≥2s) 2.0
retake_mode What to replace: replace_audio, replace_video, or replace_audio_and_video replace_audio_and_video

Extend inputs

Input Description Default
video Input video to extend (required)
duration Seconds to add (max 20) 6
extend_mode Where to add: start or end end

Output

MP4 video file with synchronized audio.

Prompting tips

  • Write one continuous paragraph in present tense
  • Be specific about camera angles, lighting, and movement
  • Include atmospheric details: lighting quality, weather, textures
  • Describe actions with active verbs: “walks”, “pans slowly”, “zooms in”
  • Longer, more detailed prompts produce better results

Also available

  • LTX 2.3 Fast — faster generation, ideal for rapid prototyping and high-volume production
Model created