LTX-2.3 Unified: Video/Image/Audio → Video

Unified pipeline for video-to-video (IC-LoRA), image-to-video, and audio-conditioned generation with LTX-2.3 — use any combination of inputs. [model] [code]

🖼️ Input Image (I2V — first frame)

🎬 Reference Video (V2V)

Video Preprocessing

Strips appearance from video → style comes from image/prompt instead

🔊 Input Audio (A2V — lipsync / BGM)

Prompt

Describe the desired output — be as detailed as possible

Duration (seconds)

1 10

V2V Conditioning Strength

How closely to follow the reference video

0 1

Enhance Prompt

High Resolution

Use Audio from Video

Extract and use the audio track from the reference video

Generated Video