Click here to try online first:

Workflow: Text-to-Lip-Sync Video – Speaking/Singing – LTX2.3 Text-to-Digital Human – No Subtitles – Module Optimization

Demo URL: https://www.runninghub.cn/post/2038618886479814658/?inviteCode=rh-v1401

Workflow: LTX2.3 – Fully Automated Prompts – Text-to-Video

Demo URL: https://www.runninghub.cn/post/2031218445026594817/?inviteCode=rh-v1401

Workflow: AA-FlashVSR Video Upscaling + RIFE Frame Interpolation

Demo URL: https://www.runninghub.cn/post/2008534784552738818/?inviteCode=rh-v1401

Workflow: Lip-Sync Speaking/Singing – LTX2.3 Image-to-Digital Human – Auto-Expand – Module Optimization – No Subtitles

Demo URL: https://www.runninghub.cn/post/2038618856104665090/?inviteCode=rh-v1401

Workflow: LTX2.3 – Fully Automated Prompts – Image-to-Video – Modular Optimized Edition

Demo URL: https://www.runninghub.cn/post/2031218459471777794/?inviteCode=rh-v1401

Workflow: LTX2.3 Fully Automated Prompts – First/Middle/Last Frame Three-Image-to-Video

Demo URL: https://www.runninghub.cn/post/2035325465820405761/?inviteCode=rh-v1401

English:

A complete ComfyUI workflow for LTX 2.3. Generates music MVs from text + audio. Supports lip-sync (speech/singing, Chinese/English). Includes Huang Chan LoRA for consistent ancient-style faces. Built-in dual sampler reduces on-screen subtitles. Auto prompt expansion via LLM node. Optimized for 10s clips at 1280x720. Also works for text-to-video. Upload your audio, set strength, and run. All models and nodes listed. Great for AI filmmakers and MV creators.

中文：

一套完整的 LTX 2.3 ComfyUI 工作流。通过文本+音频生成音乐 MV，支持对口型（说话/唱歌，中英文）。内置黄婵 LoRA，保证古风角色面部一致性。双采样器有效减少画面字幕。可通过 LLM 节点自动扩写提示词。针对 10 秒片段和 1280x720 分辨率优化。同时支持普通文生视频。上传音频、设置 LoRA 强度即可运行。所有模型和节点均已列出。适合 AI 视频创作者和 MV 制作者。