LTX2.3 Multi-image Reference Major Update | Multi-image Import 10S Similarity Lock, Stable and Consistent Characters

This LTX2.3 Multi-Image Reference Video Generation Workflow supports importing multiple reference images at once. By using global and segmented prompts to control video generation, it simultaneously locks character identity and actions, ensuring natural character movements and stable subjects without drifting within a 10-second video range.

Core Features:

• Multi-Image Reference: Import multiple reference images at once, covering different characters, actions, or props, and the system automatically blends them to generate the video.
• Global Prompts: Control the overall visual style, lighting, scene rules, and character identity stability.
• Segmented Prompts: Actions, expressions, and camera changes for each time segment can be controlled individually.
• 10S Similarity Locking: Ensure high consistency between characters and reference images through similarity constraints.
• Three-Stage Sampling Refinement: Output high-definition, clear, and detail-rich videos, avoiding blurriness and clutter.
• Easy Operation: Users only need to import reference images and text prompts to generate videos, with no need for manual segmented rendering.

🎁Claim RH Coins first, then experience the workflow! Click your avatar in the upper right corner → Invitation Code → Enter [rh-v1111] to instantly receive 1,000 RH Coins, and log in daily to get another 100 RH Coins~ For more ComfyUI workflows, tutorials, and gameplay, follow our official WeChat account (AIKSK), with simultaneous updates on Douyin / Bilibili / Xiaohongshu / YouTube (AI-KSK) ✨