One-Image Cinematic Video Generation

This is an Image-to-Video application powered by LTX-2.3.

Upload a reference image and enter a text prompt. The app first generates an initial image-to-video result, then performs a second-stage upscale and refinement, ultimately delivering a clearer, more cinematic video output. The current pipeline also includes an audio branch, making it ideal for generating short videos with strong atmosphere, stage presence, and narrative feel.

Recommended usage:

• Use clear reference images with a single subject and well-defined composition.
• Focus prompts on the subject, action, camera movement, lighting, and atmosphere.
• Best suited for subtle motion, slow-motion, and emotional shots; highly complex, large-scale scene changes are not recommended at the start.
• To make results more stable, you can emphasize "consistent action, consistent subject, natural lighting and shadows, and stable framing."

Recommended input example:
Natural subject movement, slow camera push-in, consistent action, natural changes in lighting and shadow, clear image details, smooth and stable overall motion, strong cinematic feel.

🎁Claim RH Coins first, then try the workflow! Click your profile picture in the top right corner → Invitation Code → Enter 【rh-v1111】 to instantly get 1000 RH Coins. Log in daily to claim another 100 RH Coins~ For more ComfyUI workflows, tutorials, and gameplay, follow our WeChat Official Account (AIKSK). Updates are also synced on TikTok / Bilibili / Xiaohongshu / YouTube (AI-KSK) ✨