{"id":8127,"date":"2026-05-12T00:45:00","date_gmt":"2026-05-11T16:45:00","guid":{"rendered":"https:\/\/infernews.com\/blog\/?page_id=8127"},"modified":"2026-08-21T02:06:46","modified_gmt":"2026-08-20T18:06:46","slug":"ltx","status":"publish","type":"page","link":"https:\/\/infernews.com\/blog\/ltx\/","title":{"rendered":"LTX \u5c08\u696d\u5f71\u7247\u958b\u6e90\u5de5\u5177"},"content":{"rendered":"\n<table id=\"tablepress-10\" class=\"tablepress tablepress-id-10\">\n<thead>\n<tr class=\"row-1\">\n\t<th class=\"column-1\">ComfyUI-CrossViewWarp<\/th><th class=\"column-2\"><a href=\"https:\/\/github.com\/cseti007\/ComfyUI-CrossViewWarp\" target=\"_blank\">https:\/\/github.com\/cseti007\/ComfyUI-CrossViewWarp<\/a><\/th>\n<\/tr>\n<\/thead>\n<tbody class=\"row-striping row-hover\">\n<tr class=\"row-2\">\n\t<td class=\"column-1\">IC-LoRA: CrossView Warp v2<\/td><td class=\"column-2\"><a href=\"https:\/\/huggingface.co\/Cseti\/LTX2.3-22B_IC-LoRA-CrossView-Warp_v2\" target=\"_blank\">https:\/\/huggingface.co\/Cseti\/LTX2.3-22B_IC-LoRA-CrossView-Warp_v2<\/a><\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<!-- #tablepress-10 from cache -->\n\n\n\n<h2 class=\"wp-block-heading\">LTX 2.5<\/h2>\n\n\n<figure class=\"wp-block-embed-youtube wp-block-embed is-type-video is-provider-youtube wp-embed-aspect-16-9 wp-has-aspect-ratio\"><div class=\"lyte-wrapper\" title=\"LTX 2.5 \u7121\u5be9\u67e5\u6a21\u578b\u4f86\u4e86\uff01\u514d\u8cbb\u751f\u6210 AI \u5927\u7247\uff5cVS MiniMax H3\uff0c4GB VRAM \u5be6\u6e2c\uff01\" style=\"width:853px;max-width:100%;margin:5px auto;\"><div class=\"lyMe\" id=\"WYL_6YNdJJ8Se04\" itemprop=\"video\" itemscope itemtype=\"https:\/\/schema.org\/VideoObject\"><div><meta itemprop=\"thumbnailUrl\" content=\"https:\/\/infernews.com\/blog\/wp-content\/plugins\/wp-youtube-lyte\/lyteCache.php?origThumbUrl=https%3A%2F%2Fi.ytimg.com%2Fvi%2F6YNdJJ8Se04%2Fhqdefault.jpg\" \/><meta itemprop=\"embedURL\" content=\"https:\/\/www.youtube.com\/embed\/6YNdJJ8Se04\" \/><meta itemprop=\"duration\" content=\"PT10M20S\" \/><meta itemprop=\"uploadDate\" content=\"2026-08-20T11:00:38Z\" \/><\/div><meta itemprop=\"accessibilityFeature\" content=\"captions\" \/><div id=\"lyte_6YNdJJ8Se04\" data-src=\"https:\/\/infernews.com\/blog\/wp-content\/plugins\/wp-youtube-lyte\/lyteCache.php?origThumbUrl=https%3A%2F%2Fi.ytimg.com%2Fvi%2F6YNdJJ8Se04%2Fhqdefault.jpg\" class=\"pL\"><div class=\"tC\"><div class=\"tT\" itemprop=\"name\">LTX 2.5 \u7121\u5be9\u67e5\u6a21\u578b\u4f86\u4e86\uff01\u514d\u8cbb\u751f\u6210 AI \u5927\u7247\uff5cVS MiniMax H3\uff0c4GB VRAM \u5be6\u6e2c\uff01<\/div><\/div><button tabindex=\"0\" class=\"play\"><\/button><div class=\"ctrl\"><div class=\"Lctrl\"><\/div><div class=\"Rctrl\"><\/div><\/div><\/div><noscript><a href=\"https:\/\/youtu.be\/6YNdJJ8Se04\" rel=\"nofollow\"><img loading=\"lazy\" decoding=\"async\" src=\"https:\/\/infernews.com\/blog\/wp-content\/plugins\/wp-youtube-lyte\/lyteCache.php?origThumbUrl=https%3A%2F%2Fi.ytimg.com%2Fvi%2F6YNdJJ8Se04%2F0.jpg\" alt=\"LTX 2.5 \u7121\u5be9\u67e5\u6a21\u578b\u4f86\u4e86\uff01\u514d\u8cbb\u751f\u6210 AI \u5927\u7247\uff5cVS MiniMax H3\uff0c4GB VRAM \u5be6\u6e2c\uff01\" width=\"853\" height=\"460\" \/><br \/>Watch this video on YouTube<\/a><\/noscript><meta itemprop=\"description\" content=\"LTX 2.5 \u525b\u525b\u767c\u5e03\uff01\u9019\u6b21\uff0c\u6211\u5011\u4f86\u5b8c\u6574\u9ad4\u9a57\u9019\u500b\u5f37\u5927\u7684\u958b\u6e90 AI \u8996\u983b\u6a21\u578b\uff0c\u4e26\u76f4\u63a5\u6311\u6230 MiniMax H3\uff0c\u770b\u770b\u5169\u8005\u5230\u5e95\u8ab0\u66f4\u5f37\uff01 \u8a31\u591a\u4eba\u8a8d\u70ba\u9019\u985e AI \u8996\u983b\u6a21\u578b\u9700\u8981\u8f03\u9ad8\u7684\u986f\u5b58\u624d\u80fd\u904b\u884c\uff0c\u800c\u76ee\u524d\u516c\u958b\u7684\u4f4e\u986f\u5b58\u6e2c\u8a66\u4e5f\u4e3b\u8981\u96c6\u4e2d\u5728 12GB VRAM \u4ee5\u4e0a\u3002 \u4f46\u9019\u6b21\u6211\u76f4\u63a5\u4f7f\u7528 4GB VRAM \u7684 GPU \u9032\u884c\u5be6\u6e2c\uff0c\u7d50\u679c\u767c\u73fe LTX 2.5 \u548c MiniMax H3 \u90fd\u53ef\u4ee5\u5728\u4f4e\u986f\u5b58\u74b0\u5883\u4e0b\u904b\u884c\u3002\u4e0d\u904e\uff0c\u53d7\u9650\u65bc\u986f\u5b58\u5bb9\u91cf\uff0c\u5be6\u969b\u751f\u6210\u901f\u5ea6\u6703\u6bd4\u8f03\u6162\uff0c\u9700\u8981\u66f4\u591a\u7b49\u5f85\u6642\u9593\u3002 \u793e\u7fa4\u4e5f\u6709\u4e00\u4e9b\u91cf\u5316\u6a21\u578b\u548c CPU\/RAM Offloading \u7684\u65b9\u6848\uff0c\u53ef\u4ee5\u9032\u4e00\u6b65\u964d\u4f4e GPU VRAM \u7684\u9700\u6c42\u3002LTX 2.5 \u53ef\u4ee5\u4f7f\u7528\u91cf\u5316\u6a21\u578b\uff0c\u4f8b\u5982 GGUF\uff0c\u642d\u914d CPU\/RAM Offloading\uff0c\u628a\u90e8\u5206\u6a21\u578b\u5f9e GPU \u986f\u5b58\u79fb\u5230\u7cfb\u7d71\u8a18\u61b6\u9ad4\u3002MiniMax H3 \u4e5f\u662f\u985e\u4f3c\u7684\u65b9\u6cd5\uff0c\u901a\u904e\u91cf\u5316\u6a21\u578b\u548c CPU\/RAM Offloading\uff0c\u964d\u4f4e GPU \u986f\u5b58\u7684\u9700\u6c42\u3002\u4e0d\u904e\u9019\u4e9b\u65b9\u6cd5\u6211\u5728\u672c\u671f\u8996\u983b\u4e2d\u6c92\u6709\u9032\u884c\u6e2c\u8a66\u3002 \ud83d\udc4d \u5982\u679c\u9019\u671f\u5f71\u7247\u5c0d\u4f60\u6709\u5e6b\u52a9\uff0c\u5225\u5fd8\u4e86\u9ede\u8b9a\u3001\u8a02\u95b1\u983b\u9053\u4e26\u958b\u555f\u901a\u77e5\uff0c\u652f\u6301\u6211\u7e7c\u7e8c\u6e2c\u8a66\u66f4\u591a AI \u6a21\u578b\uff01 #LTX25 #MiniMaxH3 #ComfyUI #AIVideo #AI\u5f71\u7247 #\u958b\u6e90AI \u23f1\ufe0f \u8996\u983b\u7ae0\u7bc0 00:00 LTX 2.5 \u65b0\u529f\u80fd\u4ecb\u7d39 02:32 ComfyUI \u5b89\u88dd\u8207\u57f7\u884c 03:13 Text to Video\uff5c\u6587\u5b57\u751f\u6210\u5f71\u7247 05:31 Image to Video\uff5c\u5716\u7247\u751f\u6210\u5f71\u7247 06:06 First Frame \u2192 Last Frame\uff5c\u9996\u5c3e\u5e40\u751f\u6210 06:49 LTX 2 LoRA\uff5c\u5728 LTX 2.5 \u4e2d\u4f7f\u7528 LoRA 08:51 LTX 2.5 vs MiniMax H3\uff5c\u5be6\u6e2c\u6bd4\u8f03 Minimax H3: https:\/\/youtu.be\/DvbF1wF8OuE Comfy workflow: https:\/\/docs.comfy.org\/tutorials\/video\/ltx\/ltx-2-5 LTX models: https:\/\/huggingface.co\/Lightricks\/LTX-2.5 Minimax + LTX upscaler: https:\/\/github.com\/peterducan-hub\/PeterDuncan_Comfyui\/blob\/main\/MINIMAX_H3_LTX2.5_Upscaler_v1.json LTX Director: LTX Director: https:\/\/github.com\/WhatDreamsCost\/WhatDreamsCost-ComfyUI #LTX25 #MiniMaxH3 #ComfyUI\"><\/div><\/div><div class=\"lL\" style=\"max-width:100%;width:853px;margin:5px auto;\"><\/div><figcaption><\/figcaption><\/figure>\n\n\n<h2 class=\"wp-block-heading\">\u5feb\u901f\u4e0a\u624b<\/h2>\n\n\n<figure class=\"wp-block-embed-youtube wp-block-embed is-type-video is-provider-youtube wp-embed-aspect-16-9 wp-has-aspect-ratio\"><div class=\"lyte-wrapper\" title=\"LTX-2.3 Quickstart in ComfyUI: Install, Models, and Your First Workflow\" style=\"width:853px;max-width:100%;margin:5px auto;\"><div class=\"lyMe\" id=\"WYL__YGc2Tf8lcY\" itemprop=\"video\" itemscope itemtype=\"https:\/\/schema.org\/VideoObject\"><div><meta itemprop=\"thumbnailUrl\" content=\"https:\/\/infernews.com\/blog\/wp-content\/plugins\/wp-youtube-lyte\/lyteCache.php?origThumbUrl=https%3A%2F%2Fi.ytimg.com%2Fvi%2F_YGc2Tf8lcY%2Fhqdefault.jpg\" \/><meta itemprop=\"embedURL\" content=\"https:\/\/www.youtube.com\/embed\/_YGc2Tf8lcY\" \/><meta itemprop=\"duration\" content=\"PT14M17S\" \/><meta itemprop=\"uploadDate\" content=\"2026-07-08T13:31:37Z\" \/><\/div><div id=\"lyte__YGc2Tf8lcY\" data-src=\"https:\/\/infernews.com\/blog\/wp-content\/plugins\/wp-youtube-lyte\/lyteCache.php?origThumbUrl=https%3A%2F%2Fi.ytimg.com%2Fvi%2F_YGc2Tf8lcY%2Fhqdefault.jpg\" class=\"pL\"><div class=\"tC\"><div class=\"tT\" itemprop=\"name\">LTX-2.3 Quickstart in ComfyUI: Install, Models, and Your First Workflow<\/div><\/div><button tabindex=\"0\" class=\"play\"><\/button><div class=\"ctrl\"><div class=\"Lctrl\"><\/div><div class=\"Rctrl\"><\/div><\/div><\/div><noscript><a href=\"https:\/\/youtu.be\/_YGc2Tf8lcY\" rel=\"nofollow\"><img loading=\"lazy\" decoding=\"async\" src=\"https:\/\/infernews.com\/blog\/wp-content\/plugins\/wp-youtube-lyte\/lyteCache.php?origThumbUrl=https%3A%2F%2Fi.ytimg.com%2Fvi%2F_YGc2Tf8lcY%2F0.jpg\" alt=\"LTX-2.3 Quickstart in ComfyUI: Install, Models, and Your First Workflow\" width=\"853\" height=\"460\" \/><br \/>Watch this video on YouTube<\/a><\/noscript><meta itemprop=\"description\" content=\"LTX Quickstart, Part 1: ComfyUI Install and First Workflow In part one of the LTX quickstart, we walk through installing the LTX-2.3 custom node for ComfyUI from source. You&#039;ll see how to git clone the repository, run UV sync with locked dependencies, activate the environment, and load an example workflow from the 2.3 folder. The session also covers where to find LTX-2.3 checkpoints and LoRAs on HuggingFace, how to download the Gemma 3 text encoder, where each model file belongs inside your ComfyUI directory, VRAM requirements for local generation, and how to fix common workflow issues like the bypass toggle for image-to-video. This is for developers, ComfyUI users, and technical artists comfortable with Python, git, and the command line who want to run LTX-2.3 locally on their own machine. 00:00 - 00:42 Intro: LTX Quickstart in ComfyUI 00:42 - 01:07 Who This Video Is For 01:07 - 01:34 Getting ComfyUI Desktop Installed 01:34 - 02:24 Two Install Paths: LTX Desktop vs Git Method 02:24 - 03:25 Git Clone the LTX Video Custom Node 03:25 - 03:51 Why UV Sync Frozen Matters 03:51 - 04:47 Terminal Walkthrough on Windows 04:47 - 05:39 Activating the Environment and Modifying the Code 05:39 - 06:25 Example Workflows in the 2.3 Directory 06:25 - 07:25 Loading the Two-Stage Distilled Workflow 07:25 - 08:17 Downloading Models from HuggingFace and Gemma 3 08:17 - 09:05 Where Checkpoints and LoRAs Belong 09:05 - 09:57 Hardware and VRAM Requirements 09:57 - 10:37 The Rig Running This Tutorial 10:37 - 11:31 Opening the Workflow and Fixing the Bypass Toggle 11:31 - 13:00 Pointing Model Paths for LTX-2.3, LoRA, and Gemma 13:00 - 13:23 Loading the Base Image for Image-to-Video 13:23 - 14:16 Recap and What&#039;s Next in Part 2 \ud83d\udd37 Join our Discord Community: https:\/\/discord.com\/invite\/ltxplatform \ud83d\udd37 Follow us on X: https:\/\/x.com\/ltx_io \ud83d\udd37 Follow us on Instagram: https:\/\/www.instagram.com\/ltx.io\/ About LTX LTX builds open world models: AI that understands and generates physical reality across video, audio, and multimodal inputs. Open, efficient, and yours to own. Run it on your own hardware and fine-tune it on your own IP, with 18M+ downloads and no vendor lock-in. \ud83d\udd14 Subscribe for more LTX model releases, open-source tools, training guides, and developer resources.\"><\/div><\/div><div class=\"lL\" style=\"max-width:100%;width:853px;margin:5px auto;\"><\/div><figcaption><\/figcaption><\/figure>\n\n<figure class=\"wp-block-embed-youtube wp-block-embed is-type-video is-provider-youtube wp-embed-aspect-16-9 wp-has-aspect-ratio\"><div class=\"lyte-wrapper\" title=\"LTX-2.3 Quickstart Part 2: Prompt Structure, Rendering, and Debugging in ComfyUI\" style=\"width:853px;max-width:100%;margin:5px auto;\"><div class=\"lyMe\" id=\"WYL_6Ijwv2SH11I\" itemprop=\"video\" itemscope itemtype=\"https:\/\/schema.org\/VideoObject\"><div><meta itemprop=\"thumbnailUrl\" content=\"https:\/\/infernews.com\/blog\/wp-content\/plugins\/wp-youtube-lyte\/lyteCache.php?origThumbUrl=https%3A%2F%2Fi.ytimg.com%2Fvi%2F6Ijwv2SH11I%2Fhqdefault.jpg\" \/><meta itemprop=\"embedURL\" content=\"https:\/\/www.youtube.com\/embed\/6Ijwv2SH11I\" \/><meta itemprop=\"duration\" content=\"PT13M45S\" \/><meta itemprop=\"uploadDate\" content=\"2026-07-08T13:31:42Z\" \/><\/div><div id=\"lyte_6Ijwv2SH11I\" data-src=\"https:\/\/infernews.com\/blog\/wp-content\/plugins\/wp-youtube-lyte\/lyteCache.php?origThumbUrl=https%3A%2F%2Fi.ytimg.com%2Fvi%2F6Ijwv2SH11I%2Fhqdefault.jpg\" class=\"pL\"><div class=\"tC\"><div class=\"tT\" itemprop=\"name\">LTX-2.3 Quickstart Part 2: Prompt Structure, Rendering, and Debugging in ComfyUI<\/div><\/div><button tabindex=\"0\" class=\"play\"><\/button><div class=\"ctrl\"><div class=\"Lctrl\"><\/div><div class=\"Rctrl\"><\/div><\/div><\/div><noscript><a href=\"https:\/\/youtu.be\/6Ijwv2SH11I\" rel=\"nofollow\"><img loading=\"lazy\" decoding=\"async\" src=\"https:\/\/infernews.com\/blog\/wp-content\/plugins\/wp-youtube-lyte\/lyteCache.php?origThumbUrl=https%3A%2F%2Fi.ytimg.com%2Fvi%2F6Ijwv2SH11I%2F0.jpg\" alt=\"LTX-2.3 Quickstart Part 2: Prompt Structure, Rendering, and Debugging in ComfyUI\" width=\"853\" height=\"460\" \/><br \/>Watch this video on YouTube<\/a><\/noscript><meta itemprop=\"description\" content=\"LTX Quickstart, Part 2: Prompting, Rendering, and Debugging In part two of the LTX quickstart, we finish the image-to-video workflow you set up in part one. You&#039;ll see how LTX prompts are structured (subject, action, setting, camera move, audio cues), how to use an LLM with the LTX prompting guide to expand a simple idea into a production prompt, and how to configure FPS, frame count, and video size before hitting run. The session also covers reading the workflow graph left to right, how the two-stage distilled workflow runs low-resolution generation then upscales to 1920x1088, why the first render is slower than the ones that follow, and how to debug common ComfyUI issues using the console, node reloads, and UV sync frozen for dependency mismatches. This is for developers, ComfyUI users, and technical artists who finished part one and want to generate their first LTX-2.3 video locally with synchronized sound effects. 00:00 - 00:24 Intro: Welcome Back to Part 2 00:24 - 01:19 Opening the Image-to-Video Workflow 01:19 - 02:02 Reading the Workflow Graph Left to Right 02:02 - 02:46 LTX Prompt Structure: Subject, Action, Setting, Camera, Audio 02:46 - 03:27 Using the LTX Prompting Guide with an LLM 03:27 - 04:36 Building the Pinball Prompt 04:36 - 04:55 Sound Effect Cues Inside the Prompt 04:55 - 05:37 FPS, Frame Count, and Video Size 05:37 - 06:00 Latent Video Node Settings 06:00 - 06:30 Hitting Run and Hardware Expectations 06:30 - 07:00 First Run Load Times 07:00 - 07:51 Two-Stage Workflow: Low Res Then Upscale 07:51 - 08:25 Reviewing the Final Video with Synced Audio 08:25 - 08:47 Gotcha: Restart ComfyUI After New Nodes 08:47 - 09:30 Debugging with the ComfyUI Console 09:30 - 10:10 Using an LLM to Read Console Errors 10:10 - 10:51 Dependency Mismatches and UV Sync Frozen 10:51 - 11:22 Where to Go Next: Prompting and Image-to-Video Guides 11:22 - 12:15 LTX Trainer and the GitHub Docs 12:15 - 12:38 Kudos to the LTX Documentation 12:38 - 13:22 Join the LTX Discord Community 13:22 - 13:44 Final Thoughts &amp; Thanks \ud83d\udd37 Join our Discord Community: https:\/\/discord.com\/invite\/ltxplatform \ud83d\udd37 Follow us on X: https:\/\/x.com\/ltx_io \ud83d\udd37 Follow us on Instagram: https:\/\/www.instagram.com\/ltx.io\/ About LTX LTX builds open world models: AI that understands and generates physical reality across video, audio, and multimodal inputs. Open, efficient, and yours to own. Run it on your own hardware and fine-tune it on your own IP, with 18M+ downloads and no vendor lock-in. \ud83d\udd14 Subscribe for more LTX model releases, open-source tools, training guides, and developer resources.\"><\/div><\/div><div class=\"lL\" style=\"max-width:100%;width:853px;margin:5px auto;\"><\/div><figcaption><\/figcaption><\/figure>\n\n\n<hr class=\"wp-block-separator has-alpha-channel-opacity\"\/>\n\n\n\n<h2 class=\"wp-block-heading\">All in one<\/h2>\n\n\n<figure class=\"wp-block-embed-youtube wp-block-embed is-type-video is-provider-youtube wp-embed-aspect-16-9 wp-has-aspect-ratio\"><div class=\"lyte-wrapper\" title=\"How to Enhance Videos with LTX 2.3 | Remove Watermarks &amp;amp; Text, Upscale &amp;amp; Outpaint\" style=\"width:853px;max-width:100%;margin:5px auto;\"><div class=\"lyMe\" id=\"WYL_bAN7XdRa-20\" itemprop=\"video\" itemscope itemtype=\"https:\/\/schema.org\/VideoObject\"><div><meta itemprop=\"thumbnailUrl\" content=\"https:\/\/infernews.com\/blog\/wp-content\/plugins\/wp-youtube-lyte\/lyteCache.php?origThumbUrl=https%3A%2F%2Fi.ytimg.com%2Fvi%2FbAN7XdRa-20%2Fhqdefault.jpg\" \/><meta itemprop=\"embedURL\" content=\"https:\/\/www.youtube.com\/embed\/bAN7XdRa-20\" \/><meta itemprop=\"duration\" content=\"PT4M26S\" \/><meta itemprop=\"uploadDate\" content=\"2026-07-06T12:15:17Z\" \/><\/div><meta itemprop=\"accessibilityFeature\" content=\"captions\" \/><div id=\"lyte_bAN7XdRa-20\" data-src=\"https:\/\/infernews.com\/blog\/wp-content\/plugins\/wp-youtube-lyte\/lyteCache.php?origThumbUrl=https%3A%2F%2Fi.ytimg.com%2Fvi%2FbAN7XdRa-20%2Fhqdefault.jpg\" class=\"pL\"><div class=\"tC\"><div class=\"tT\" itemprop=\"name\">How to Enhance Videos with LTX 2.3 | Remove Watermarks &amp; Text, Upscale &amp; Outpaint<\/div><\/div><button tabindex=\"0\" class=\"play\"><\/button><div class=\"ctrl\"><div class=\"Lctrl\"><\/div><div class=\"Rctrl\"><\/div><\/div><\/div><noscript><a href=\"https:\/\/youtu.be\/bAN7XdRa-20\" rel=\"nofollow\"><img loading=\"lazy\" decoding=\"async\" src=\"https:\/\/infernews.com\/blog\/wp-content\/plugins\/wp-youtube-lyte\/lyteCache.php?origThumbUrl=https%3A%2F%2Fi.ytimg.com%2Fvi%2FbAN7XdRa-20%2F0.jpg\" alt=\"How to Enhance Videos with LTX 2.3 | Remove Watermarks &amp;amp; Text, Upscale &amp;amp; Outpaint\" width=\"853\" height=\"460\" \/><br \/>Watch this video on YouTube<\/a><\/noscript><meta itemprop=\"description\" content=\"Looking for an easy way to remove watermarks, subtitles, restore blurry videos, or even convert vertical videos into horizontal ones\u2014all inside a single ComfyUI workflow? In this tutorial, I&#039;ll show you my latest LTX 2.3 All-in-One Video Enhancement Workflow, designed to handle multiple video editing tasks with a single sampler while keeping VRAM usage as low as possible. In this video, you&#039;ll learn how to: \u2022 Remove watermarks, logos, icons, and subtitles from videos \u2022 Restore and upscale low-quality footage to HD \u2022 Expand videos horizontally or vertically with AI Outpainting \u2022 Configure the workflow for the best balance between quality and performance Unlike the traditional LTX 2.3 pipeline, this workflow bypasses the standard downscale\/upscale process to preserve as much detail as possible. Although it takes a little longer to render, the improvement in visual quality is well worth it. \ud83d\udcac If you have any questions or ideas for future workflows, feel free to leave a comment below. I always enjoy hearing your feedback. If you find this video helpful, don&#039;t forget to Like, Subscribe, and turn on notifications so you won&#039;t miss future ComfyUI workflows, AI video tutorials, and workflow updates. \u2728 All workflows and resources are shared for free. If you\u2019d like to support the channel, you can buy me a coffee \u2764\ufe0f: https:\/\/ko-fi.com\/fututek -------------------------------------------------- \u2728Run on Cloud GPU: https:\/\/runpod.io?ref=99cuxxw1 -------------------------------------------------- \u2728Workflow and resources: https:\/\/drive.google.com\/drive\/folders\/1Eby6vrdr3dyINW7AaigGy6WxTMw3AAj9?usp=sharing - LTX2.3 Model (Place in \/ComfyUI\/models\/diffusion_models) https:\/\/huggingface.co\/vantagewithai\/LTX-2.3-GGUF\/blob\/main\/dev\/ltx-2-3-22b-dev-Q4_K_M.gguf - Text Encoder (Place in \/ComfyUI\/models\/text_encoders) https:\/\/huggingface.co\/Comfy-Org\/ltx-2\/resolve\/main\/split_files\/text_encoders\/gemma_3_12B_it_fp8_scaled.safetensors https:\/\/huggingface.co\/Kijai\/LTX2.3_comfy\/blob\/main\/text_encoders\/ltx-2.3_text_projection_bf16.safetensors - VAE (Place in \/ComfyUI\/models\/vae) https:\/\/huggingface.co\/Kijai\/LTX2.3_comfy\/blob\/main\/vae\/LTX23_audio_vae_bf16.safetensors https:\/\/huggingface.co\/Kijai\/LTX2.3_comfy\/blob\/main\/vae\/LTX23_video_vae_bf16.safetensors - Distilled Loras (Place in \/ComfyUI\/models\/loras) https:\/\/huggingface.co\/Kijai\/LTX2.3_comfy\/blob\/main\/loras\/ltx-2.3-22b-distilled-lora-dynamic_fro09_avg_rank_105_bf16.safetensors - Enhancement Loras (Place in \/ComfyUI\/models\/loras) https:\/\/huggingface.co\/joyfox\/LTX2.3-ICEdit-Insight\/resolve\/main\/ltx2.3-ic-subtitles-remove-general.safetensors?download=true https:\/\/huggingface.co\/joyfox\/LTX2.3-ICEdit-Insight\/resolve\/main\/ltx2.3-ic-video-upscale-general.safetensors?download=true https:\/\/huggingface.co\/joyfox\/LTX2.3-ICEdit-Insight\/resolve\/main\/ltx2.3-ic-watermark-remove-general.safetensors?download=true https:\/\/huggingface.co\/joyfox\/LTX2.3-ICEdit-Insight\/resolve\/main\/ltx2.3-video-restoration-general.safetensors?download=true https:\/\/huggingface.co\/oumoumad\/LTX-2.3-22b-IC-LoRA-Outpaint\/resolve\/main\/ltx-2.3-22b-ic-lora-outpaint.safetensors?download=true https:\/\/huggingface.co\/Lightricks\/LTX-2-19b-IC-LoRA-Detailer\/resolve\/main\/ltx-2-19b-ic-lora-detailer.safetensors\"><\/div><\/div><div class=\"lL\" style=\"max-width:100%;width:853px;margin:5px auto;\"><\/div><figcaption><\/figcaption><\/figure>\n\n<figure class=\"wp-block-embed-youtube wp-block-embed is-type-video is-provider-youtube wp-embed-aspect-16-9 wp-has-aspect-ratio\"><div class=\"lyte-wrapper\" title=\"The Best IC-LoRA for Reference Sheet Video Generation &mdash; LTX 2.3 in ComfyUI\" style=\"width:853px;max-width:100%;margin:5px auto;\"><div class=\"lyMe\" id=\"WYL_jNWZL2azMjs\" itemprop=\"video\" itemscope itemtype=\"https:\/\/schema.org\/VideoObject\"><div><meta itemprop=\"thumbnailUrl\" content=\"https:\/\/infernews.com\/blog\/wp-content\/plugins\/wp-youtube-lyte\/lyteCache.php?origThumbUrl=https%3A%2F%2Fi.ytimg.com%2Fvi%2FjNWZL2azMjs%2Fhqdefault.jpg\" \/><meta itemprop=\"embedURL\" content=\"https:\/\/www.youtube.com\/embed\/jNWZL2azMjs\" \/><meta itemprop=\"duration\" content=\"PT12M56S\" \/><meta itemprop=\"uploadDate\" content=\"2026-07-03T13:52:16Z\" \/><\/div><div id=\"lyte_jNWZL2azMjs\" data-src=\"https:\/\/infernews.com\/blog\/wp-content\/plugins\/wp-youtube-lyte\/lyteCache.php?origThumbUrl=https%3A%2F%2Fi.ytimg.com%2Fvi%2FjNWZL2azMjs%2Fhqdefault.jpg\" class=\"pL\"><div class=\"tC\"><div class=\"tT\" itemprop=\"name\">The Best IC-LoRA for Reference Sheet Video Generation \u2014 LTX 2.3 in ComfyUI<\/div><\/div><button tabindex=\"0\" class=\"play\"><\/button><div class=\"ctrl\"><div class=\"Lctrl\"><\/div><div class=\"Rctrl\"><\/div><\/div><\/div><noscript><a href=\"https:\/\/youtu.be\/jNWZL2azMjs\" rel=\"nofollow\"><img loading=\"lazy\" decoding=\"async\" src=\"https:\/\/infernews.com\/blog\/wp-content\/plugins\/wp-youtube-lyte\/lyteCache.php?origThumbUrl=https%3A%2F%2Fi.ytimg.com%2Fvi%2FjNWZL2azMjs%2F0.jpg\" alt=\"The Best IC-LoRA for Reference Sheet Video Generation &mdash; LTX 2.3 in ComfyUI\" width=\"853\" height=\"460\" \/><br \/>Watch this video on YouTube<\/a><\/noscript><meta itemprop=\"description\" content=\"Workflow- https:\/\/buymeacoffee.com\/ai_jigyasa\/e\/553772 Paid Workflow with local Asset (reference sheet Generator)- https:\/\/buymeacoffee.com\/ai_jigyasa\/e\/554470 In this video, I&#039;ll show you how to generate AI videos directly from a reference sheet using LTX 2.3 and IC-LoRA inside ComfyUI \u2014 no character setup, no separate image-to-video step, just drop in your reference sheet and let the model do the work. This workflow tests and compares the Ingredients IC-LoRA by Lightricks and the MSR (Multiple Subject Reference) IC-LoRA by LiconStudio side by side, so you can see exactly which one holds character consistency better for your use case. Featuring IC-LoRA Ingredients: https:\/\/huggingface.co\/Lightricks\/LTX-2.3-22b-IC-LoRA-Ingredients Featuring IC-LoRA MSR: https:\/\/huggingface.co\/LiconStudio\/LTX-2.3-Multiple-Subject-Reference What you&#039;ll learn: LTX 2.3 Audio-Video Model IC-LoRA Reference Editing ComfyUI Workflow &amp; Resource Monitoring LoRA Comparison (Ingredients vs. Lykon) Refinement &amp; Enhancer LoRAs Technical Generation Settings Character Sheet Prompting Automated Prompt Generation Model links Models (dev) BF16: https:\/\/huggingface.co\/Lightricks\/LTX-2.3\/resolve\/main\/ltx-2.3-22b-dev.safetensors?download=true FP8: https:\/\/huggingface.co\/Kijai\/LTX2.3_comfy\/resolve\/main\/diffusion_models\/ltx-2.3-22b-dev_transformer_only_fp8_scaled.safetensors?download=true GGUF: https:\/\/huggingface.co\/QuantStack\/LTX-2.3-GGUF\/tree\/main Models (distilled) BF16: https:\/\/huggingface.co\/Lightricks\/LTX-2.3\/resolve\/main\/ltx-2.3-22b-distilled-1.1.safetensors?download=true FP8: https:\/\/huggingface.co\/Kijai\/LTX2.3_comfy\/resolve\/main\/diffusion_models\/ltx-2.3-22b-distilled_transformer_only_fp8_scaled.safetensors?download=true GGUF: https:\/\/huggingface.co\/QuantStack\/LTX-2.3-GGUF\/tree\/main text_encoders https:\/\/huggingface.co\/Comfy-Org\/ltx-2\/resolve\/main\/split_files\/text_encoders\/gemma_3_12B_it.safetensors?download=true https:\/\/huggingface.co\/Kijai\/LTX2.3_comfy\/resolve\/main\/text_encoders\/ltx-2.3_text_projection_bf16.safetensors?download=true vae https:\/\/huggingface.co\/Kijai\/LTX2.3_comfy\/resolve\/main\/vae\/LTX23_video_vae_bf16.safetensors?download=true https:\/\/huggingface.co\/Kijai\/LTX2.3_comfy\/resolve\/main\/vae\/LTX23_audio_vae_bf16.safetensors?download=true loras (distilled) https:\/\/huggingface.co\/Lightricks\/LTX-2.3\/resolve\/main\/ltx-2.3-22b-distilled-lora-384-1.1.safetensors?download=true IC-LoRAs \ud83d\udd39 Ingredients (by Lightricks) \u2014 Reference sheet \u2192 video consistency More info: https:\/\/huggingface.co\/Lightricks\/LTX-2.3-22b-IC-LoRA-Ingredients https:\/\/huggingface.co\/Lightricks\/LTX-2.3-22b-IC-LoRA-Ingredients\/resolve\/main\/ltx-2.3-22b-ic-lora-ingredients-0.9.safetensors?download=true \ud83d\udd39 MSR \u2014 Multiple Subject Reference (by LiconStudio) \u2014 Multi-character reference consistency More info: https:\/\/huggingface.co\/LiconStudio\/LTX-2.3-Multiple-Subject-Reference V1: https:\/\/huggingface.co\/LiconStudio\/LTX-2.3-Multiple-Subject-Reference\/resolve\/main\/LTX-2.3-Licon-MSR-V1.safetensors?download=true Test Version: https:\/\/huggingface.co\/LiconStudio\/LTX-2.3-Multiple-Subject-Reference\/resolve\/main\/LTX2.3-Licon-MSR-test_version.safetensors?download=true \ud83d\udd39 Union Control (by Lightricks) \u2014 Multi-control reference https:\/\/huggingface.co\/Lightricks\/LTX-2.3-22b-IC-LoRA-Union-Control\/resolve\/main\/ltx-2.3-22b-ic-lora-union-control-ref0.5.safetensors?download=true latent_upscale_models https:\/\/huggingface.co\/Lightricks\/LTX-2.3\/resolve\/main\/ltx-2.3-spatial-upscaler-x2-1.1.safetensors?download=true Vocals Extraction https:\/\/huggingface.co\/Kijai\/MelBandRoFormer_comfy\/resolve\/main\/MelBandRoformer_fp32.safetensors?download=true \ud83d\udc4d Support the Channel If you found this helpful: \ud83d\udc4d Like the video \ud83d\udd14 Subscribe for more AI workflows \ud83d\udcac Comment what you want to see next #LTX23 #ComfyUI #AIVideo #GenerativeAI #ICLoRA #ReferenceSheet #CharacterConsistency #Ref2Edit #LightricksAI #OpenSourceAI #AIContentCreation #ComfyUIWorkflow\"><\/div><\/div><div class=\"lL\" style=\"max-width:100%;width:853px;margin:5px auto;\"><\/div><figcaption><\/figcaption><\/figure>\n\n<figure class=\"wp-block-embed-youtube wp-block-embed is-type-video is-provider-youtube wp-embed-aspect-16-9 wp-has-aspect-ratio\"><div class=\"lyte-wrapper\" title=\"JoyAI ComfyUI: Edit Multiple Images at Once (+ Free Workflow)\" style=\"width:853px;max-width:100%;margin:5px auto;\"><div class=\"lyMe\" id=\"WYL_hp203XBlVd0\" itemprop=\"video\" itemscope itemtype=\"https:\/\/schema.org\/VideoObject\"><div><meta itemprop=\"thumbnailUrl\" content=\"https:\/\/infernews.com\/blog\/wp-content\/plugins\/wp-youtube-lyte\/lyteCache.php?origThumbUrl=https%3A%2F%2Fi.ytimg.com%2Fvi%2Fhp203XBlVd0%2Fhqdefault.jpg\" \/><meta itemprop=\"embedURL\" content=\"https:\/\/www.youtube.com\/embed\/hp203XBlVd0\" \/><meta itemprop=\"duration\" content=\"PT15M11S\" \/><meta itemprop=\"uploadDate\" content=\"2026-07-19T12:30:21Z\" \/><\/div><div id=\"lyte_hp203XBlVd0\" data-src=\"https:\/\/infernews.com\/blog\/wp-content\/plugins\/wp-youtube-lyte\/lyteCache.php?origThumbUrl=https%3A%2F%2Fi.ytimg.com%2Fvi%2Fhp203XBlVd0%2Fhqdefault.jpg\" class=\"pL\"><div class=\"tC\"><div class=\"tT\" itemprop=\"name\">JoyAI ComfyUI: Edit Multiple Images at Once (+ Free Workflow)<\/div><\/div><button tabindex=\"0\" class=\"play\"><\/button><div class=\"ctrl\"><div class=\"Lctrl\"><\/div><div class=\"Rctrl\"><\/div><\/div><\/div><noscript><a href=\"https:\/\/youtu.be\/hp203XBlVd0\" rel=\"nofollow\"><img loading=\"lazy\" decoding=\"async\" src=\"https:\/\/infernews.com\/blog\/wp-content\/plugins\/wp-youtube-lyte\/lyteCache.php?origThumbUrl=https%3A%2F%2Fi.ytimg.com%2Fvi%2Fhp203XBlVd0%2F0.jpg\" alt=\"JoyAI ComfyUI: Edit Multiple Images at Once (+ Free Workflow)\" width=\"853\" height=\"460\" \/><br \/>Watch this video on YouTube<\/a><\/noscript><meta itemprop=\"description\" content=\"Learn how to set up JoyAI Image Edit in ComfyUI to edit your images, change clothes, and outpaint vertical photos. You can do all of this without changing the face or the original scene. You can download our free workflow below! In this tutorial, we build the complete ComfyUI workflow for both JoyAI Image Edit and JoyAI Image Edit Plus models. You will see how to load the Qwen-VL 8B text encoder. We also use its INT8 version to save your memory. You will also load the Wan 2.1 VAE model file. This video covers advanced local editing and swapping characters. We also show you a white-padding workflow. This method solves the vertical cropping issue completely. It keeps the face consistent. \u2b07\ufe0f Resources &amp; Links: JoyAI ComfyUI Workflow JSON: https:\/\/aistudynow.com\/how-to-set-up-joyai-image-edit-in-comfyui-no-more-face-drift\/ JoyAI Image Edit Models &amp; Text Encoder: Models https:\/\/huggingface.co\/jdopensource\/JoyAI-Image-Edit-Plus-ComfyUI\/tree\/main\/diffusion_models https:\/\/huggingface.co\/jdopensource\/JoyAI-Image-Edit-Plus-ComfyUI\/blob\/main\/diffusion_models\/joyai_image_edit_plus_int8_convrot.safetensors Plus https:\/\/huggingface.co\/jdopensource\/JoyAI-Image-Edit-Plus-ComfyUI\/blob\/main\/diffusion_models\/joyai_image_edit_plus_bf16.safetensors https:\/\/huggingface.co\/jdopensource\/JoyAI-Image-Edit-Plus-ComfyUI\/blob\/main\/diffusion_models\/joyai_image_edit_plus_int8_convrot.safetensors Text Encoder: https:\/\/huggingface.co\/jdopensource\/JoyAI-Image-Edit-ComfyUI\/blob\/main\/text_encoders\/qwen3vl_8b_joyimage_edit_bf16.safetensors https:\/\/huggingface.co\/jdopensource\/JoyAI-Image-Edit-ComfyUI\/blob\/main\/text_encoders\/qwen3vl_8b_joyimage_edit_int8_convrot.safetensors Plus https:\/\/huggingface.co\/jdopensource\/JoyAI-Image-Edit-Plus-ComfyUI\/blob\/main\/text_encoders\/qwen3vl_8b_joyimage_edit_plus_bf16.safetensors https:\/\/huggingface.co\/jdopensource\/JoyAI-Image-Edit-Plus-ComfyUI\/blob\/main\/diffusion_models\/joyai_image_edit_plus_int8_convrot.safetensors Wan 2.1 VAE Safetensors: https:\/\/huggingface.co\/jdopensource\/JoyAI-Image-Edit-ComfyUI\/blob\/main\/vae\/wan_2.1_vae.safetensors Wan2.1-VAE-upscale2x : https:\/\/huggingface.co\/spacepxl\/Wan2.1-VAE-upscale2x\/blob\/main\/Wan2.1_VAE_upscale2x_imageonly_real_v1.safetensors If this tutorial helped you get cleaner local edits, subscribe so you don&#039;t miss our next ComfyUI advanced guide! #ComfyUI #JoyAI #ImageEditing\"><\/div><\/div><div class=\"lL\" style=\"max-width:100%;width:853px;margin:5px auto;\"><\/div><figcaption><\/figcaption><\/figure>\n\n\n<h2 class=\"wp-block-heading\">MSR<\/h2>\n\n\n<figure class=\"wp-block-embed-youtube wp-block-embed is-type-video is-provider-youtube wp-embed-aspect-16-9 wp-has-aspect-ratio\"><div class=\"lyte-wrapper\" title=\"LTX 2.3 MSR V2 Update - Single Image to Multiple Scenes Workflow &amp;amp; Review\" style=\"width:853px;max-width:100%;margin:5px auto;\"><div class=\"lyMe\" id=\"WYL_EFm4z0ZF20M\" itemprop=\"video\" itemscope itemtype=\"https:\/\/schema.org\/VideoObject\"><div><meta itemprop=\"thumbnailUrl\" content=\"https:\/\/infernews.com\/blog\/wp-content\/plugins\/wp-youtube-lyte\/lyteCache.php?origThumbUrl=https%3A%2F%2Fi.ytimg.com%2Fvi%2FEFm4z0ZF20M%2Fhqdefault.jpg\" \/><meta itemprop=\"embedURL\" content=\"https:\/\/www.youtube.com\/embed\/EFm4z0ZF20M\" \/><meta itemprop=\"duration\" content=\"PT15M22S\" \/><meta itemprop=\"uploadDate\" content=\"2026-07-25T13:00:22Z\" \/><\/div><div id=\"lyte_EFm4z0ZF20M\" data-src=\"https:\/\/infernews.com\/blog\/wp-content\/plugins\/wp-youtube-lyte\/lyteCache.php?origThumbUrl=https%3A%2F%2Fi.ytimg.com%2Fvi%2FEFm4z0ZF20M%2Fhqdefault.jpg\" class=\"pL\"><div class=\"tC\"><div class=\"tT\" itemprop=\"name\">LTX 2.3 MSR V2 Update - Single Image to Multiple Scenes Workflow &amp; Review<\/div><\/div><button tabindex=\"0\" class=\"play\"><\/button><div class=\"ctrl\"><div class=\"Lctrl\"><\/div><div class=\"Rctrl\"><\/div><\/div><\/div><noscript><a href=\"https:\/\/youtu.be\/EFm4z0ZF20M\" rel=\"nofollow\"><img loading=\"lazy\" decoding=\"async\" src=\"https:\/\/infernews.com\/blog\/wp-content\/plugins\/wp-youtube-lyte\/lyteCache.php?origThumbUrl=https%3A%2F%2Fi.ytimg.com%2Fvi%2FEFm4z0ZF20M%2F0.jpg\" alt=\"LTX 2.3 MSR V2 Update - Single Image to Multiple Scenes Workflow &amp;amp; Review\" width=\"853\" height=\"460\" \/><br \/>Watch this video on YouTube<\/a><\/noscript><meta itemprop=\"description\" content=\"This video covers the MSR V2 (Multi-Subject Reference version 2) LoRA model for LTX Video 2.3, an update from Licon Studio that improves character consistency, object stability, and scene logic in reference-to-video generation. The V2 model is trained to maintain outfit styles, accessories, and body type across multiple scenes and camera angles \u2014 about 80 percent style consistency with a single reference image. We compare V1 vs V2 side by side, test close-up vs non-close-up reference images, and demonstrate the results with a music video demo. This is for ComfyUI users who work with LTX 2.3 and want to generate multiple scenes from a single character reference image. If you have been using reference-to-video but struggling with flickering, inconsistent outfits, or characters that disappear mid-clip, MSR V2 addresses those issues. The workflow uses the LiconMSR custom node for character sheet input, Prompt Relay for text prompts, and NAG for guided generation. It works with the same workflow as V1 \u2014 just swap the LoRA file. The key finding: MSR V2 excels at outfit and style consistency (about 80 percent) but facial recognition remains the limitation of all reference-to-video approaches. For exact character face identity across scenes, training a character LoRA with Krea 2 or similar tools is still the best path. Reference-to-video is a quick way to mimic overall character style, not a replacement for dedicated character LoRA training. Resources: Attached Freebie workflow here for the LTX 2.3 22B + MSR V2 LoRA + LiconMSR : https:\/\/www.patreon.com\/aifuturetech\/posts\/ltx-2-3-msr-v2-164795767?utm_source=youtube&amp;utm_medium=video&amp;utm_campaign=20260725 LiconStudio\/LTX-2.3-Multiple-Subject-Reference by @liconstudio https:\/\/huggingface.co\/LiconStudio\/LTX-2.3-Multiple-Subject-Reference -------------------------------------------------------------------------------------------------------------------------------- Local Workstation GPU : https:\/\/amzn.to\/3XfXsAO -------------------------------------------------------------------------------------------------------------------------------- If You Like tutorial like this, You Can Support Our Work In Patreon: https:\/\/www.patreon.com\/c\/aifuturetech #comfyui #ltx2 #aivideoediting #referneceTovideo\"><\/div><\/div><div class=\"lL\" style=\"max-width:100%;width:853px;margin:5px auto;\"><\/div><figcaption><\/figcaption><\/figure>\n\n<figure class=\"wp-block-embed-youtube wp-block-embed is-type-video is-provider-youtube wp-embed-aspect-16-9 wp-has-aspect-ratio\"><div class=\"lyte-wrapper\" title=\"I Made Elon Musk Do This... (Free Open-Source AI Video)\" style=\"width:853px;max-width:100%;margin:5px auto;\"><div class=\"lyMe\" id=\"WYL_AtEstFHy3Dc\" itemprop=\"video\" itemscope itemtype=\"https:\/\/schema.org\/VideoObject\"><div><meta itemprop=\"thumbnailUrl\" content=\"https:\/\/infernews.com\/blog\/wp-content\/plugins\/wp-youtube-lyte\/lyteCache.php?origThumbUrl=https%3A%2F%2Fi.ytimg.com%2Fvi%2FAtEstFHy3Dc%2Fhqdefault.jpg\" \/><meta itemprop=\"embedURL\" content=\"https:\/\/www.youtube.com\/embed\/AtEstFHy3Dc\" \/><meta itemprop=\"duration\" content=\"PT10M9S\" \/><meta itemprop=\"uploadDate\" content=\"2026-06-20T02:25:31Z\" \/><\/div><div id=\"lyte_AtEstFHy3Dc\" data-src=\"https:\/\/infernews.com\/blog\/wp-content\/plugins\/wp-youtube-lyte\/lyteCache.php?origThumbUrl=https%3A%2F%2Fi.ytimg.com%2Fvi%2FAtEstFHy3Dc%2Fhqdefault.jpg\" class=\"pL\"><div class=\"tC\"><div class=\"tT\" itemprop=\"name\">I Made Elon Musk Do This... (Free Open-Source AI Video)<\/div><\/div><button tabindex=\"0\" class=\"play\"><\/button><div class=\"ctrl\"><div class=\"Lctrl\"><\/div><div class=\"Rctrl\"><\/div><\/div><\/div><noscript><a href=\"https:\/\/youtu.be\/AtEstFHy3Dc\" rel=\"nofollow\"><img loading=\"lazy\" decoding=\"async\" src=\"https:\/\/infernews.com\/blog\/wp-content\/plugins\/wp-youtube-lyte\/lyteCache.php?origThumbUrl=https%3A%2F%2Fi.ytimg.com%2Fvi%2FAtEstFHy3Dc%2F0.jpg\" alt=\"I Made Elon Musk Do This... (Free Open-Source AI Video)\" width=\"853\" height=\"460\" \/><br \/>Watch this video on YouTube<\/a><\/noscript><meta itemprop=\"description\" content=\"Can open-source AI video finally handle flawless character consistency? Today, we are stress-testing the brand new LTX 2.3 Multiple Subject Reference (MSR) model by LiconStudio. I put this 22-billion parameter architecture to the absolute test using multiple characters, custom props, and chaotic prompt changes to see if it actually lives up to the developer hype or falls apart under pressure. Best of all, I\u2019ve put together a 100% free Kaggle notebook so you can run this model yourself without needing a high-end local GPU. We\u2019ll break down the token-level visual memory system, look at the best settings to prevent your generations from breaking, and look at the brutal truth regarding its identity retention limitations. \u25bd\u00a0TOOLS &amp; RESOURCES MENTIONED\u00a0\u25bd \u25ba My Free Notebooks Collections: https:\/\/github.com\/TeamAIQ\/Colab-notebooks \u25ba LTX 2.3 MSR Hugging Face Page: https:\/\/huggingface.co\/LiconStudio\/LTX-2.3-Multiple-Subject-Reference \u25ac\u25ac\u25ac\u25ac\u25ac\u25ac\u25ac\u25ac\u25ac\u25ac\u25ac\u25ac\u25ac\u25ac\u25ac\u25ac\u25ac\u25ac\u25ac\u25ac\u25ac \ud83e\udd1d Business: TeamAIQ@yahoo.com \ud83d\ude80 X \/ Twitter: https:\/\/x.com\/AiquestAcademy \u2615 My bio link: https:\/\/aiquest.site\/ \u25ac\u25ac\u25ac\u25ac\u25ac\u25ac\u25ac\u25ac\u25ac\u25ac\u25ac\u25ac\u25ac\u25ac\u25ac\u25ac\u25ac\u25ac\u25ac\u25ac\u25ac #AIVideo #OpenSourceAI #CharacterConsistency\"><\/div><\/div><div class=\"lL\" style=\"max-width:100%;width:853px;margin:5px auto;\"><\/div><figcaption><\/figcaption><\/figure>\n\n\n<h2 class=\"wp-block-heading\">LTX Trainer<\/h2>\n\n\n<figure class=\"wp-block-embed-youtube wp-block-embed is-type-video is-provider-youtube wp-embed-aspect-16-9 wp-has-aspect-ratio\"><div class=\"lyte-wrapper\" title=\"LTX Trainer: Open-Source LoRA Training\" style=\"width:853px;max-width:100%;margin:5px auto;\"><div class=\"lyMe\" id=\"WYL_7N2kTJQCgEk\" itemprop=\"video\" itemscope itemtype=\"https:\/\/schema.org\/VideoObject\"><div><meta itemprop=\"thumbnailUrl\" content=\"https:\/\/infernews.com\/blog\/wp-content\/plugins\/wp-youtube-lyte\/lyteCache.php?origThumbUrl=https%3A%2F%2Fi.ytimg.com%2Fvi%2F7N2kTJQCgEk%2Fhqdefault.jpg\" \/><meta itemprop=\"embedURL\" content=\"https:\/\/www.youtube.com\/embed\/7N2kTJQCgEk\" \/><meta itemprop=\"duration\" content=\"PT1M21S\" \/><meta itemprop=\"uploadDate\" content=\"2026-06-17T14:41:08Z\" \/><\/div><div id=\"lyte_7N2kTJQCgEk\" data-src=\"https:\/\/infernews.com\/blog\/wp-content\/plugins\/wp-youtube-lyte\/lyteCache.php?origThumbUrl=https%3A%2F%2Fi.ytimg.com%2Fvi%2F7N2kTJQCgEk%2Fhqdefault.jpg\" class=\"pL\"><div class=\"tC\"><div class=\"tT\" itemprop=\"name\">LTX Trainer: Open-Source LoRA Training<\/div><\/div><button tabindex=\"0\" class=\"play\"><\/button><div class=\"ctrl\"><div class=\"Lctrl\"><\/div><div class=\"Rctrl\"><\/div><\/div><\/div><noscript><a href=\"https:\/\/youtu.be\/7N2kTJQCgEk\" rel=\"nofollow\"><img loading=\"lazy\" decoding=\"async\" src=\"https:\/\/infernews.com\/blog\/wp-content\/plugins\/wp-youtube-lyte\/lyteCache.php?origThumbUrl=https%3A%2F%2Fi.ytimg.com%2Fvi%2F7N2kTJQCgEk%2F0.jpg\" alt=\"LTX Trainer: Open-Source LoRA Training\" width=\"853\" height=\"460\" \/><br \/>Watch this video on YouTube<\/a><\/noscript><meta itemprop=\"description\" content=\"LTX Trainer is here. Train LTX on your own data. LTX Trainer is the open-source framework for customizing LTX with your own characters, styles, workflows, and IP. Build LoRAs and IC-LoRAs so the model learns your world and reproduces it consistently across projects. Your data and IP never leave your infrastructure. What&#039;s new in this release: \u25cf New conditioning modes out of the box: video, audio, cross-modal, and reference (IC-LoRAs) \u25cf Combine multiple conditioning methods from a single framework \u25cf Agentic setup: use Claude or any LLM to configure your training runs \u25cf Fully open source, maintained by the core LTX team Get started: GitHub: https:\/\/github.com\/Lightricks\/LTX-2\/blob\/main\/packages\/ltx-trainer\/README.md Documentation: https:\/\/docs.ltx.video\/open-source-model\/getting-started\/overview HuggingFace: https:\/\/huggingface.co\/collections\/Lightricks\/ltx-23-creative-lab \ud83d\udd14 Subscribe for updates on LTX model releases, open-source tools, and developer resources.\"><\/div><\/div><div class=\"lL\" style=\"max-width:100%;width:853px;margin:5px auto;\"><\/div><figcaption><\/figcaption><\/figure>\n\n<figure class=\"wp-block-embed-youtube wp-block-embed is-type-video is-provider-youtube wp-embed-aspect-16-9 wp-has-aspect-ratio\"><div class=\"lyte-wrapper\" title=\"LTX Trainer Explained - Ingredients IC-LoRA The NEW Reference-to-Video Model Explained\" style=\"width:853px;max-width:100%;margin:5px auto;\"><div class=\"lyMe\" id=\"WYL_7nMTLCfohqs\" itemprop=\"video\" itemscope itemtype=\"https:\/\/schema.org\/VideoObject\"><div><meta itemprop=\"thumbnailUrl\" content=\"https:\/\/infernews.com\/blog\/wp-content\/plugins\/wp-youtube-lyte\/lyteCache.php?origThumbUrl=https%3A%2F%2Fi.ytimg.com%2Fvi%2F7nMTLCfohqs%2Fhqdefault.jpg\" \/><meta itemprop=\"embedURL\" content=\"https:\/\/www.youtube.com\/embed\/7nMTLCfohqs\" \/><meta itemprop=\"duration\" content=\"PT7M5S\" \/><meta itemprop=\"uploadDate\" content=\"2026-06-19T19:08:03Z\" \/><\/div><div id=\"lyte_7nMTLCfohqs\" data-src=\"https:\/\/infernews.com\/blog\/wp-content\/plugins\/wp-youtube-lyte\/lyteCache.php?origThumbUrl=https%3A%2F%2Fi.ytimg.com%2Fvi%2F7nMTLCfohqs%2Fhqdefault.jpg\" class=\"pL\"><div class=\"tC\"><div class=\"tT\" itemprop=\"name\">LTX Trainer Explained - Ingredients IC-LoRA The NEW Reference-to-Video Model Explained<\/div><\/div><button tabindex=\"0\" class=\"play\"><\/button><div class=\"ctrl\"><div class=\"Lctrl\"><\/div><div class=\"Rctrl\"><\/div><\/div><\/div><noscript><a href=\"https:\/\/youtu.be\/7nMTLCfohqs\" rel=\"nofollow\"><img loading=\"lazy\" decoding=\"async\" src=\"https:\/\/infernews.com\/blog\/wp-content\/plugins\/wp-youtube-lyte\/lyteCache.php?origThumbUrl=https%3A%2F%2Fi.ytimg.com%2Fvi%2F7nMTLCfohqs%2F0.jpg\" alt=\"LTX Trainer Explained - Ingredients IC-LoRA The NEW Reference-to-Video Model Explained\" width=\"853\" height=\"460\" \/><br \/>Watch this video on YouTube<\/a><\/noscript><meta itemprop=\"description\" content=\"LTX-2 AI video model, LTX Trainer fine-tuning, Ingredients IC-LoRA, consistent AI video characters, open-source AI video generation, and custom LoRA training just got a massive upgrade. Lightricks released LTX 2.3 with a fully overhauled open-source trainer that lets you fine-tune a 22 billion parameter multimodal AI model that generates synchronized video and audio in one single pass \u2014 and you own the weights. In this video we break down how the LTX Trainer works, the difference between LoRA, IC-LoRA, and full fine-tuning, how the Ingredients reference-to-video model solves the AI character consistency problem using a simple reference grid image, and how to automate the entire pipeline using AI agents, ComfyUI, and Gemma 3. This video is for AI filmmakers, creative directors, indie developers, and content creators who want to build hyper-consistent AI video with full control over their models without being locked into expensive cloud platforms. ------------------------------------------------------------------- ETSY Store: https:\/\/www.etsy.com\/shop\/maxonshire Twitter: https:\/\/twitter.com\/MaxonShire ------------------------------------------------------------------- Relevant hashtags: #LTX #LTXTrainer #AIVideo #LoRA #LTX2 #ConsistentAICharacters #OpenSourceAI #AIVideoGeneration #Lightricks #ImageToVideo #TextToVideo #AIAnimation #AI\"><\/div><\/div><div class=\"lL\" style=\"max-width:100%;width:853px;margin:5px auto;\"><\/div><figcaption><\/figcaption><\/figure>\n\n\n<hr class=\"wp-block-separator has-alpha-channel-opacity\"\/>\n\n\n\n<h2 class=\"wp-block-heading\">LTX Director<\/h2>\n\n\n<figure class=\"wp-block-embed-youtube wp-block-embed is-type-video is-provider-youtube wp-embed-aspect-16-9 wp-has-aspect-ratio\"><div class=\"lyte-wrapper\" title=\"LTX 2.3 Director Prompt Generator: Auto-Create Timed Prompts\" style=\"width:853px;max-width:100%;margin:5px auto;\"><div class=\"lyMe\" id=\"WYL_C79pI_I1jgU\" itemprop=\"video\" itemscope itemtype=\"https:\/\/schema.org\/VideoObject\"><div><meta itemprop=\"thumbnailUrl\" content=\"https:\/\/infernews.com\/blog\/wp-content\/plugins\/wp-youtube-lyte\/lyteCache.php?origThumbUrl=https%3A%2F%2Fi.ytimg.com%2Fvi%2FC79pI_I1jgU%2Fhqdefault.jpg\" \/><meta itemprop=\"embedURL\" content=\"https:\/\/www.youtube.com\/embed\/C79pI_I1jgU\" \/><meta itemprop=\"duration\" content=\"PT7M39S\" \/><meta itemprop=\"uploadDate\" content=\"2026-07-21T16:59:46Z\" \/><\/div><div id=\"lyte_C79pI_I1jgU\" data-src=\"https:\/\/infernews.com\/blog\/wp-content\/plugins\/wp-youtube-lyte\/lyteCache.php?origThumbUrl=https%3A%2F%2Fi.ytimg.com%2Fvi%2FC79pI_I1jgU%2Fhqdefault.jpg\" class=\"pL\"><div class=\"tC\"><div class=\"tT\" itemprop=\"name\">LTX 2.3 Director Prompt Generator: Auto-Create Timed Prompts<\/div><\/div><button tabindex=\"0\" class=\"play\"><\/button><div class=\"ctrl\"><div class=\"Lctrl\"><\/div><div class=\"Rctrl\"><\/div><\/div><\/div><noscript><a href=\"https:\/\/youtu.be\/C79pI_I1jgU\" rel=\"nofollow\"><img loading=\"lazy\" decoding=\"async\" src=\"https:\/\/infernews.com\/blog\/wp-content\/plugins\/wp-youtube-lyte\/lyteCache.php?origThumbUrl=https%3A%2F%2Fi.ytimg.com%2Fvi%2FC79pI_I1jgU%2F0.jpg\" alt=\"LTX 2.3 Director Prompt Generator: Auto-Create Timed Prompts\" width=\"853\" height=\"460\" \/><br \/>Watch this video on YouTube<\/a><\/noscript><meta itemprop=\"description\" content=\"ChatLLM by Abacus AI: https:\/\/chatllm.abacus.ai\/tra Open the LTX 2.3 Director Prompt Studio: https:\/\/ltx-director-prompt-studio.abacusai.app\/ Download the LTX 2.3 Director ComfyUI workflow: https:\/\/drive.google.com\/file\/d\/1r9PEJW-8JXfqyS6i6R15kuX_iPa3qiwZ\/view?usp=drive_link LTX 2.3 Director Prompt Generator transforms a simple scene idea into a Global Prompt, precisely timed Segment Prompts, natural dialogue, contextual sound design, and a validated timeline for the LTX 2.3 Director workflow in ComfyUI. In this tutorial, I show the complete process behind LTX 2.3 Director Prompt Studio\u2014from building the application with the Agent inside ChatLLM by Abacus AI to generating structured prompts and using them in a local ComfyUI workflow. The app converts each chronological action into one timed Director segment while keeping persistent scene details inside a separate Global Prompt. It automatically organizes dialogue, environmental audio, camera direction, visual style, and continuity requirements without forcing the user to write a massive technical prompt. You\u2019ll see the Prompt Studio create a complete 20-second dialogue timeline from five short emotional actions, generate natural spoken lines, add continuous beach ambience, validate every time range, and prepare the output for LTX 2.3 Director. I also use the same tool to structure an atmospheric horse-and-aurora scene without dialogue, demonstrating how the timeline can handle environmental motion, evolving expressions, and automatic sound direction. The current public release is focused on Text-to-Video and includes: - one Global Prompt for clip-wide consistency - one timed Segment Prompt for every action - automatic contextual sound design - separated and correctly formatted dialogue - deterministic timeline allocation - gap and overlap detection - camera and continuity validation - individual and complete-timeline copy controls Sponsored by Abacus.AI. ChatLLM by Abacus AI brings leading language models, creative tools, and AI Agents into one workspace. The standard subscription is &#36;10 per month after any introductory offer. The final generation may vary depending on the prompt, model, seed, resolution, and workflow settings. The Prompt Studio is designed to make the timeline faster to create, inspect, copy, and refine. Subscribe to Tensor Alchemist for practical ComfyUI tutorials, local AI video workflows, low-VRAM tools, and production-focused generative AI experiments. \ud83d\udccc Timestamps: 0:00 - Prompt Studio Generation 0:20 - Five Simple Action Lines 0:37 - Horse and Aurora Scene 0:47 - LTX 2.3 Director Prompting 1:28 - LTX 2.3 Prompt Studio 2:17 - Building the App with the Abacus Agent 3:50 - Prompt Studio Tutorial \ud83d\udcbb My Setup : Laptop: Gigabyte A16 GPU: NVIDIA RTX 5060 (8GB VRAM) RAM: DDR5 32GB (Dual Channel) \ud83d\udcbc Business Inquiries &amp; Sponsorships For brand partnerships, hardware evaluations, software reviews, or enterprise tool placements targeting an engaged audience of local machine learning developers, AI researchers, and digital media professionals: please contact me via email. #ltx23 #comfyui #aivideo #ltxdirector #aivideogeneration #localai\"><\/div><\/div><div class=\"lL\" style=\"max-width:100%;width:853px;margin:5px auto;\"><\/div><figcaption><\/figcaption><\/figure>\n\n<figure class=\"wp-block-embed-youtube wp-block-embed is-type-video is-provider-youtube wp-embed-aspect-16-9 wp-has-aspect-ratio\"><div class=\"lyte-wrapper\" title=\"LTX Director 2.0 Is INSANE &mdash; Edit Existing Videos in ComfyUI\" style=\"width:853px;max-width:100%;margin:5px auto;\"><div class=\"lyMe\" id=\"WYL_BFE9v9LZpIU\" itemprop=\"video\" itemscope itemtype=\"https:\/\/schema.org\/VideoObject\"><div><meta itemprop=\"thumbnailUrl\" content=\"https:\/\/infernews.com\/blog\/wp-content\/plugins\/wp-youtube-lyte\/lyteCache.php?origThumbUrl=https%3A%2F%2Fi.ytimg.com%2Fvi%2FBFE9v9LZpIU%2Fhqdefault.jpg\" \/><meta itemprop=\"embedURL\" content=\"https:\/\/www.youtube.com\/embed\/BFE9v9LZpIU\" \/><meta itemprop=\"duration\" content=\"PT11M21S\" \/><meta itemprop=\"uploadDate\" content=\"2026-06-26T16:03:57Z\" \/><\/div><div id=\"lyte_BFE9v9LZpIU\" data-src=\"https:\/\/infernews.com\/blog\/wp-content\/plugins\/wp-youtube-lyte\/lyteCache.php?origThumbUrl=https%3A%2F%2Fi.ytimg.com%2Fvi%2FBFE9v9LZpIU%2Fhqdefault.jpg\" class=\"pL\"><div class=\"tC\"><div class=\"tT\" itemprop=\"name\">LTX Director 2.0 Is INSANE \u2014 Edit Existing Videos in ComfyUI<\/div><\/div><button tabindex=\"0\" class=\"play\"><\/button><div class=\"ctrl\"><div class=\"Lctrl\"><\/div><div class=\"Rctrl\"><\/div><\/div><\/div><noscript><a href=\"https:\/\/youtu.be\/BFE9v9LZpIU\" rel=\"nofollow\"><img loading=\"lazy\" decoding=\"async\" src=\"https:\/\/infernews.com\/blog\/wp-content\/plugins\/wp-youtube-lyte\/lyteCache.php?origThumbUrl=https%3A%2F%2Fi.ytimg.com%2Fvi%2FBFE9v9LZpIU%2F0.jpg\" alt=\"LTX Director 2.0 Is INSANE &mdash; Edit Existing Videos in ComfyUI\" width=\"853\" height=\"460\" \/><br \/>Watch this video on YouTube<\/a><\/noscript><meta itemprop=\"description\" content=\"LTX Director 2.0 is finally starting to feel like a real AI video editor inside ComfyUI. In this video, I show how to use the new LTX Director 2.0 workflow to import an existing video, work with the timeline, select a specific fragment, use Retake Mode, and change only the part of the video you want \u2014 without regenerating the entire scene from scratch. We will test several practical examples: * adding lightning behind the characters * replacing coffee cups with bananas * changing outfits into dresses * turning night into day * using local and global prompts correctly * understanding when InPaint is needed * using Edit Anything LoRA for stronger object and clothing changes * preparing the workflow for Full HD generation inside ComfyUI This is not just text-to-video generation anymore. This is the beginning of controlled AI video editing directly inside ComfyUI. Free workflow, model links, and useful materials: https:\/\/www.patreon.com\/AntiMatterComfy\/posts\/ltx-director-2-0-162147683?utm_medium=clipboard_copy&amp;utm_source=copyLink&amp;utm_campaign=postshare_creator&amp;utm_content=join_link In the next video, I will go deeper into IC Video \/ IC-LoRA and how to transfer motion, pose, or style from a reference video into your own character or scene. If you are interested in ComfyUI, LTX 2.3, LTX Director, Flux, local AI video generation, and AI filmmaking \u2014 subscribe to the channel. #LTXDirector #ComfyUI #LTX23 #AIVideo #AIFilmmaking #RetakeMode #LocalAI #ComfyUIWorkflow #AIAnimation #Flux\"><\/div><\/div><div class=\"lL\" style=\"max-width:100%;width:853px;margin:5px auto;\"><\/div><figcaption><\/figcaption><\/figure>\n\n<figure class=\"wp-block-embed-youtube wp-block-embed is-type-video is-provider-youtube wp-embed-aspect-16-9 wp-has-aspect-ratio\"><div class=\"lyte-wrapper\" title=\"LTX Director + 4 Panel + Global Ref | Local AI Video Workflow\" style=\"width:853px;max-width:100%;margin:5px auto;\"><div class=\"lyMe\" id=\"WYL_rdWmc1BWhKc\" itemprop=\"video\" itemscope itemtype=\"https:\/\/schema.org\/VideoObject\"><div><meta itemprop=\"thumbnailUrl\" content=\"https:\/\/infernews.com\/blog\/wp-content\/plugins\/wp-youtube-lyte\/lyteCache.php?origThumbUrl=https%3A%2F%2Fi.ytimg.com%2Fvi%2FrdWmc1BWhKc%2Fhqdefault.jpg\" \/><meta itemprop=\"embedURL\" content=\"https:\/\/www.youtube.com\/embed\/rdWmc1BWhKc\" \/><meta itemprop=\"duration\" content=\"PT5M\" \/><meta itemprop=\"uploadDate\" content=\"2026-06-07T14:05:05Z\" \/><\/div><meta itemprop=\"accessibilityFeature\" content=\"captions\" \/><div id=\"lyte_rdWmc1BWhKc\" data-src=\"https:\/\/infernews.com\/blog\/wp-content\/plugins\/wp-youtube-lyte\/lyteCache.php?origThumbUrl=https%3A%2F%2Fi.ytimg.com%2Fvi%2FrdWmc1BWhKc%2Fhqdefault.jpg\" class=\"pL\"><div class=\"tC\"><div class=\"tT\" itemprop=\"name\">LTX Director + 4 Panel + Global Ref | Local AI Video Workflow<\/div><\/div><button tabindex=\"0\" class=\"play\"><\/button><div class=\"ctrl\"><div class=\"Lctrl\"><\/div><div class=\"Rctrl\"><\/div><\/div><\/div><noscript><a href=\"https:\/\/youtu.be\/rdWmc1BWhKc\" rel=\"nofollow\"><img loading=\"lazy\" decoding=\"async\" src=\"https:\/\/infernews.com\/blog\/wp-content\/plugins\/wp-youtube-lyte\/lyteCache.php?origThumbUrl=https%3A%2F%2Fi.ytimg.com%2Fvi%2FrdWmc1BWhKc%2F0.jpg\" alt=\"LTX Director + 4 Panel + Global Ref | Local AI Video Workflow\" width=\"853\" height=\"460\" \/><br \/>Watch this video on YouTube<\/a><\/noscript><meta itemprop=\"description\" content=\"Camera lab github link: https:\/\/github.com\/ai2764\/Camera-lab In this video, I show my extended LTX Director setup inside Camera Lab. This feature combines a 4-panel storyboard method, LTX Director-style shot control, and a global reference image system to make local AI video generation easier to direct shot by shot. I also added dependency notes so a coding agent can help pull the project, follow the setup guide, and get it running locally. Credits: Inspired by Tudou\u2019s 4-panel storyboard idea and What Dreams Cost\u2019s LTX Director workflow. #LTXDirector #CameraLab #AIVideoGeneration #LocalAI #ComfyUI #LTXVideo #AIFilmmaking #AIDirector #4PanelStoryboard #GlobalReference #ImageToVideo #LocalVideoGeneration #AIVideoWorkflow #StoryboardToVideo #PromptRelay #DirectorTool #AITools #GenerativeVideo #VideoGenerationTutorial\"><\/div><\/div><div class=\"lL\" style=\"max-width:100%;width:853px;margin:5px auto;\"><\/div><figcaption><\/figcaption><\/figure>\n\n<figure class=\"wp-block-embed-youtube wp-block-embed is-type-video is-provider-youtube wp-embed-aspect-16-9 wp-has-aspect-ratio\"><div class=\"lyte-wrapper\" title=\"LTX Director 2 + SEED HUNTER workflow | The Dangers of Convenience\/Overhaul Nodes\" style=\"width:853px;max-width:100%;margin:5px auto;\"><div class=\"lyMe\" id=\"WYL_vGxZIS4J0Mg\" itemprop=\"video\" itemscope itemtype=\"https:\/\/schema.org\/VideoObject\"><div><meta itemprop=\"thumbnailUrl\" content=\"https:\/\/infernews.com\/blog\/wp-content\/plugins\/wp-youtube-lyte\/lyteCache.php?origThumbUrl=https%3A%2F%2Fi.ytimg.com%2Fvi%2FvGxZIS4J0Mg%2Fhqdefault.jpg\" \/><meta itemprop=\"embedURL\" content=\"https:\/\/www.youtube.com\/embed\/vGxZIS4J0Mg\" \/><meta itemprop=\"duration\" content=\"PT3M23S\" \/><meta itemprop=\"uploadDate\" content=\"2026-06-21T22:14:45Z\" \/><\/div><div id=\"lyte_vGxZIS4J0Mg\" data-src=\"https:\/\/infernews.com\/blog\/wp-content\/plugins\/wp-youtube-lyte\/lyteCache.php?origThumbUrl=https%3A%2F%2Fi.ytimg.com%2Fvi%2FvGxZIS4J0Mg%2Fhqdefault.jpg\" class=\"pL\"><div class=\"tC\"><div class=\"tT\" itemprop=\"name\">LTX Director 2 + SEED HUNTER workflow | The Dangers of Convenience\/Overhaul Nodes<\/div><\/div><button tabindex=\"0\" class=\"play\"><\/button><div class=\"ctrl\"><div class=\"Lctrl\"><\/div><div class=\"Rctrl\"><\/div><\/div><\/div><noscript><a href=\"https:\/\/youtu.be\/vGxZIS4J0Mg\" rel=\"nofollow\"><img loading=\"lazy\" decoding=\"async\" src=\"https:\/\/infernews.com\/blog\/wp-content\/plugins\/wp-youtube-lyte\/lyteCache.php?origThumbUrl=https%3A%2F%2Fi.ytimg.com%2Fvi%2FvGxZIS4J0Mg%2F0.jpg\" alt=\"LTX Director 2 + SEED HUNTER workflow | The Dangers of Convenience\/Overhaul Nodes\" width=\"853\" height=\"460\" \/><br \/>Watch this video on YouTube<\/a><\/noscript><meta itemprop=\"description\" content=\"LTX DIRECTOR 2 SEED HUNTER WF: https:\/\/civitai.com\/models\/2720649\/ltx-director-2-seed-hunter-workflow (or for non-civitAI users: https:\/\/pastes.io\/rSkkMJV7) As per my video, I recommend you use traditional workflows such as my other SEED HUNTER workflows below: Single input: https:\/\/civitai.com\/models\/2676452\/ltx-23-seed-hunter-multiroll-workflow FFLF input: https:\/\/civitai.com\/models\/2688482\/ltx-23-fflf-seed-hunter-multiroll-workflow If you appreciate my work, please jump over to Patreon and follow for free (or not!): https:\/\/www.patreon.com\/c\/foxfuressence\"><\/div><\/div><div class=\"lL\" style=\"max-width:100%;width:853px;margin:5px auto;\"><\/div><figcaption><\/figcaption><\/figure>\n\n<figure class=\"wp-block-embed-youtube wp-block-embed is-type-video is-provider-youtube wp-embed-aspect-16-9 wp-has-aspect-ratio\"><div class=\"lyte-wrapper\" title=\"\u591a\u5e27\u53c2\u8003+\u63d0\u793a\u8bcd\u63a5\u529b\uff1a\u63ed\u79d8 LTX Director \u5982\u4f55\u5b9e\u73b0\u597d\u83b1\u575e\u7ea7\u8fd0\u955c\u63a7\u5236|\u73a9\u8f6c LTX Director \u7684 5 \u4e2a\u9ad8\u9636\u6280\u5de7\uff1a\u8ba9\u4f60\u7684 AI \u89c6\u9891\u544a\u522b\u50f5\u786c\u4e0e\u95ea\u70c1\" style=\"width:853px;max-width:100%;margin:5px auto;\"><div class=\"lyMe\" id=\"WYL_IAr8rd9q4ZE\" itemprop=\"video\" itemscope itemtype=\"https:\/\/schema.org\/VideoObject\"><div><meta itemprop=\"thumbnailUrl\" content=\"https:\/\/infernews.com\/blog\/wp-content\/plugins\/wp-youtube-lyte\/lyteCache.php?origThumbUrl=https%3A%2F%2Fi.ytimg.com%2Fvi%2FIAr8rd9q4ZE%2Fhqdefault.jpg\" \/><meta itemprop=\"embedURL\" content=\"https:\/\/www.youtube.com\/embed\/IAr8rd9q4ZE\" \/><meta itemprop=\"duration\" content=\"PT12M32S\" \/><meta itemprop=\"uploadDate\" content=\"2026-05-25T05:49:02Z\" \/><\/div><meta itemprop=\"accessibilityFeature\" content=\"captions\" \/><div id=\"lyte_IAr8rd9q4ZE\" data-src=\"https:\/\/infernews.com\/blog\/wp-content\/plugins\/wp-youtube-lyte\/lyteCache.php?origThumbUrl=https%3A%2F%2Fi.ytimg.com%2Fvi%2FIAr8rd9q4ZE%2Fhqdefault.jpg\" class=\"pL\"><div class=\"tC\"><div class=\"tT\" itemprop=\"name\">\u591a\u5e27\u53c2\u8003+\u63d0\u793a\u8bcd\u63a5\u529b\uff1a\u63ed\u79d8 LTX Director \u5982\u4f55\u5b9e\u73b0\u597d\u83b1\u575e\u7ea7\u8fd0\u955c\u63a7\u5236|\u73a9\u8f6c LTX Director \u7684 5 \u4e2a\u9ad8\u9636\u6280\u5de7\uff1a\u8ba9\u4f60\u7684 AI \u89c6\u9891\u544a\u522b\u50f5\u786c\u4e0e\u95ea\u70c1<\/div><\/div><button tabindex=\"0\" class=\"play\"><\/button><div class=\"ctrl\"><div class=\"Lctrl\"><\/div><div class=\"Rctrl\"><\/div><\/div><\/div><noscript><a href=\"https:\/\/youtu.be\/IAr8rd9q4ZE\" rel=\"nofollow\"><img loading=\"lazy\" decoding=\"async\" src=\"https:\/\/infernews.com\/blog\/wp-content\/plugins\/wp-youtube-lyte\/lyteCache.php?origThumbUrl=https%3A%2F%2Fi.ytimg.com%2Fvi%2FIAr8rd9q4ZE%2F0.jpg\" alt=\"\u591a\u5e27\u53c2\u8003+\u63d0\u793a\u8bcd\u63a5\u529b\uff1a\u63ed\u79d8 LTX Director \u5982\u4f55\u5b9e\u73b0\u597d\u83b1\u575e\u7ea7\u8fd0\u955c\u63a7\u5236|\u73a9\u8f6c LTX Director \u7684 5 \u4e2a\u9ad8\u9636\u6280\u5de7\uff1a\u8ba9\u4f60\u7684 AI \u89c6\u9891\u544a\u522b\u50f5\u786c\u4e0e\u95ea\u70c1\" width=\"853\" height=\"460\" \/><br \/>Watch this video on YouTube<\/a><\/noscript><meta itemprop=\"description\" content=\"\u4f60\u7684 AI \u89c6\u9891\u603b\u662f\u4e0d\u542c\u6307\u6325\u5417\uff1f\ud83c\udfac \u6b22\u8fce\u8ba4\u8bc6 LTX Director\uff0c\u8fd9\u662f LTX 2.3 \u4f53\u7cfb\u4e2d\u6700\u5f3a\u5927\u7684\u201c\u5168\u80fd\u578b\u201d\u8282\u70b9\uff0c\u5b83\u76f4\u63a5\u628a ComfyUI \u53d8\u6210\u4e86\u4e00\u4e2a\u4e13\u4e1a\u7684\u5bfc\u6f14\u63a7\u573a\u53f0\uff01 \u672c\u671f\u89c6\u9891\u6211\u5c06\u4e3a\u4f60\u6df1\u5ea6\u62c6\u89e3 Prompt Relay (\u63d0\u793a\u8bcd\u63a5\u529b) \u4e0e LTX Director \u7684\u6838\u5fc3\u8054\u7cfb\uff0c\u6559\u4f60\u5982\u4f55\u5c06\u4e0d\u540c\u7684\u6307\u4ee4\u7cbe\u51c6\u5206\u914d\u5230\u89c6\u9891\u7684\u4e0d\u540c\u7247\u6bb5\u3002\u6700\u91cd\u8981\u7684\u662f\uff0c\u5b98\u65b9\u63d0\u4f9b\u7684\u5de5\u4f5c\u6d41\u6548\u679c\u5f80\u5f80\u5dee\u5f3a\u4eba\u610f\u2014\u2014\u56e0\u6b64\u6211\u4eb2\u81ea\u6253\u9020\u4e86\u4e00\u5957**\u201cLTX Director \u6539\u8fdb\u7248\u5de5\u4f5c\u6d41\u201d**\uff0c\u901a\u8fc7 ManualSigmas \u91c7\u6837\u4f18\u5316\uff0c\u663e\u8457\u63d0\u5347\u4e86\u52a8\u4f5c\u5e45\u5ea6\u548c\u753b\u9762\u8d28\u611f\u3002 \ud83d\udd25 \u672c\u671f\u5e72\u8d27\u5185\u5bb9\u5168\u89e3\u6790\uff1a \u2022 \ud83e\udde0 \u5bfc\u6f14\u601d\u7ef4\u903b\u8f91\uff1a \u8be6\u89e3\u201c\u5168\u5c40\u63d0\u793a\u8bcd\u201d\u9501\u5b9a\u8eab\u4efd\u4e0e\u201c\u5c40\u90e8\u63d0\u793a\u8bcd\u201d\u5207\u6362\u52a8\u4f5c\u7684\u5e73\u8861\u672f\u3002 \u2022 \ud83d\udee0\ufe0f \u5de5\u4f5c\u6d41\u5927\u6539\u9020\uff1a \u63ed\u79d8\u5b98\u65b9\u8bbe\u7f6e\u7684\u5c40\u9650\uff0c\u6f14\u793a\u6539\u8fdb\u7248\u5982\u4f55\u8ba9\u52a8\u4f5c\u66f4\u6d41\u7545\u3001\u7ec6\u8282\u66f4\u624e\u5b9e\u3002 \u2022 \ud83c\udfad \u65e0\u5ba1\u67e5\u6a21\u578b\u8054\u52a8\uff1a \u5b9e\u6d4b\u914d\u5408 Sulphur \u6a21\u578b\uff0c\u8ba9\u4eba\u7269\u52a8\u4f5c\u548c\u955c\u5934\u4e92\u52a8\u66f4\u52a0\u5927\u80c6\u653e\u5f97\u5f00\u3002 \u2022 \ud83c\udf9e\ufe0f \u591a\u5e27\u63a7\u5236\u6280\u5de7\uff1a \u624b\u628a\u624b\u6559\u4f60\u9996\u5c3e\u5e27\u53c2\u8003\u3001\u591a\u4e2d\u95f4\u5e27\u5e03\u5c40\uff0c\u4ee5\u53ca\u5982\u4f55\u907f\u5f00\u591a\u5e27\u53c2\u8003\u5bfc\u81f4\u7684\u201c\u7ffb\u8f66\u201d\u9677\u9631\u3002 \u2022 \u23f1\ufe0f \u8282\u594f\u638c\u63a7\u529b\uff1a \u5982\u4f55\u6839\u636e\u52a8\u4f5c\u590d\u6742\u5ea6\u7cbe\u51c6\u5206\u914d\u5e27\u6570\uff0c\u5b9e\u73b0\u7535\u5f71\u7ea7\u7684\u526a\u8f91\u8282\u594f\u3002 #runninghub #LTXDirector #LTX23 #ComfyUI #AIVideo #PromptRelay #AIAnimation #VideoGeneration #AIFilmmaking #RunningHub #WhatDreamsCost #ManualSigmas #SulphurAI #AITutorial #GenerativeAI #VideoTimeline #AI\u7ed8\u753b #AI\u89c6\u9891 #\u5bfc\u6f14\u6a21\u5f0f #\u63d0\u793a\u8bcd\u63a5\u529b #\u5de5\u4f5c\u6d41\u5206\u4eab #\u4eba\u5de5\u667a\u80fd ------------------------------------------------------------------------------------------------------------------------ \u76f8\u5173\u8d44\u6e90 https:\/\/github.com\/WhatDreamsCost\/WhatDreamsCost-ComfyUI \u5728\u7ebf\u5de5\u4f5c\u6d41 1\uff1aLTX Director-i2v-prompt_replay-fixed \u4f53\u9a8c\u5730\u5740(\u56fd\u5185)\uff1ahttps:\/\/www.runninghub.cn\/post\/2057141350105178114\/?inviteCode=kol01-rh107 \u4f53\u9a8c\u5730\u5740(\u56fd\u9645)\uff1ahttps:\/\/www.runninghub.ai\/post\/2057141350105178114\/?inviteCode=kol02-rh021 \u5de5\u4f5c\u6d41\u4e0e\u63d0\u793a\u8bcd\u5de5\u7a0b\u4e0b\u8f7d https:\/\/github.com\/amao2001\/ganloss-latent-space\/blob\/main\/workflow\/2026-5-25LTXDirector-i2v-prompt_replay-fixed.json runninghub\u7c89\u4e1d\u798f\u5229 \u6ce8\u518c\u90011000\u70b9\u7b97\u529b \u56fd\u5916\u7528\u6237 https:\/\/www.runninghub.ai\/?utm_source=kol02-RH021 \u56fd\u5185\u7528\u6237 https:\/\/www.runninghub.cn\/?utm_source=kol01-RH107 ------------------------------------------------------------------------------------------------------------------------ \u80cc\u666f\u97f3\u4e50\u7531 AI \u751f\u6210\"><\/div><\/div><div class=\"lL\" style=\"max-width:100%;width:853px;margin:5px auto;\"><\/div><figcaption><\/figcaption><\/figure>\n\n\n<hr class=\"wp-block-separator has-alpha-channel-opacity\"\/>\n\n\n\n<h2 class=\"wp-block-heading\">Cross View Camera Control<\/h2>\n\n\n<figure class=\"wp-block-embed-youtube wp-block-embed is-type-video is-provider-youtube wp-embed-aspect-16-9 wp-has-aspect-ratio\"><div class=\"lyte-wrapper\" title=\"This ComfyUI Node Can Fully Control Video Camera Motion\" style=\"width:853px;max-width:100%;margin:5px auto;\"><div class=\"lyMe\" id=\"WYL_HpKGNd0I1jQ\" itemprop=\"video\" itemscope itemtype=\"https:\/\/schema.org\/VideoObject\"><div><meta itemprop=\"thumbnailUrl\" content=\"https:\/\/infernews.com\/blog\/wp-content\/plugins\/wp-youtube-lyte\/lyteCache.php?origThumbUrl=https%3A%2F%2Fi.ytimg.com%2Fvi%2FHpKGNd0I1jQ%2Fhqdefault.jpg\" \/><meta itemprop=\"embedURL\" content=\"https:\/\/www.youtube.com\/embed\/HpKGNd0I1jQ\" \/><meta itemprop=\"duration\" content=\"PT12M22S\" \/><meta itemprop=\"uploadDate\" content=\"2026-07-29T13:17:29Z\" \/><\/div><div id=\"lyte_HpKGNd0I1jQ\" data-src=\"https:\/\/infernews.com\/blog\/wp-content\/plugins\/wp-youtube-lyte\/lyteCache.php?origThumbUrl=https%3A%2F%2Fi.ytimg.com%2Fvi%2FHpKGNd0I1jQ%2Fhqdefault.jpg\" class=\"pL\"><div class=\"tC\"><div class=\"tT\" itemprop=\"name\">This ComfyUI Node Can Fully Control Video Camera Motion<\/div><\/div><button tabindex=\"0\" class=\"play\"><\/button><div class=\"ctrl\"><div class=\"Lctrl\"><\/div><div class=\"Rctrl\"><\/div><\/div><\/div><noscript><a href=\"https:\/\/youtu.be\/HpKGNd0I1jQ\" rel=\"nofollow\"><img loading=\"lazy\" decoding=\"async\" src=\"https:\/\/infernews.com\/blog\/wp-content\/plugins\/wp-youtube-lyte\/lyteCache.php?origThumbUrl=https%3A%2F%2Fi.ytimg.com%2Fvi%2FHpKGNd0I1jQ%2F0.jpg\" alt=\"This ComfyUI Node Can Fully Control Video Camera Motion\" width=\"853\" height=\"460\" \/><br \/>Watch this video on YouTube<\/a><\/noscript><meta itemprop=\"description\" content=\"This video covers the CrossView Warp IC-LoRA for LTX Video 2.3 22B \u2014 an upgrade from the previous CrossView Prompt that used text-based camera angle switching. The Warp version adds a drag-and-drop UI for full camera trajectory control: set keyframes at specific frame numbers, rotate a virtual camera sphere to set angles, and the IC-LoRA re-renders your source video from those new viewpoints. The video also tests the Cinematic VAE from Hugging Face, which swaps into the workflow with no settings changes. The CrossView Warp comes from the same author (Cseti) who created the Cameraman V2 and CrossView Prompt IC-LoRAs. This is for ComfyUI users and AI video creators who want dynamic camera control in LTX 2.3 without writing camera angle text prompts. If you have a static tracking shot and want to add orbit, pan, tilt, and multi-angle cuts, this IC-LoRA lets you do it visually. The custom node (ComfyUI CrossViewWarp) integrates with Depth Anything V2 for depth map extraction and provides a keyframe-based trajectory editor where you drag camera positions on a sphere. The CrossView Warp requires 1080p resolution (720p does not work with this IC-LoRA). The workflow uses two-pass sampling: first pass at half resolution, second pass upscales to 1080p. The Cinematic VAE adds a subtle cinematic quality improvement. Together, these components let you transform boring tracking shots into dynamic multi-angle sequences from a single source video. The freebie workflow attached below: https:\/\/www.patreon.com\/aifuturetech\/posts\/165166332?utm_source=youtube&amp;utm_medium=video&amp;utm_campaign=20260729 ComfyUI-CrossViewWarp https:\/\/github.com\/cseti007\/ComfyUI-CrossViewWarp Cseti\/LTX2.3-22B_IC-LoRA-CrossView-Warp https:\/\/huggingface.co\/Cseti\/LTX2.3-22B_IC-LoRA-CrossView-Warp rzgar\/LTX-2.3-Cinematic-VAE https:\/\/huggingface.co\/rzgar\/LTX-2.3-Cinematic-VAE 0:00 - Introduction to LTX CrossView Warp Overview of the new IC LoRA version that uses warp for dynamic camera rotations and transitions rather than just text prompts. 1:02 - Cinematic VAE Integration Introduction of the new cinematic VAE from Hugging Face and how swapping it in can enhance the visual feel of your video output. 1:26 - ComfyUI Workflow &amp; Custom Nodes Explanation of the custom &quot;Crossview Warp&quot; nodes required for this integration and how they work with the IC LoRA. 2:24 - Using the Camera Control UI Demonstration of the interactive camera control (the ball-shaped UI) for rotating 360 degrees or adjusting pitch. 2:58 - Model Loader &amp; Optimization A walkthrough of the model loader setup, including the LTX 2.3 checkpoint, distilled LoRA, and Sage Attention for optimization. 3:36 - Text Prompts &amp; Trigger Words How to use the &quot;crossview&quot; trigger word to activate the specific camera effects. 4:16 - Input Files &amp; Resolution Constraints Discussion on video input paths and resolution settings, noting that 1080p is the required default for the warp IC LoRA. 4:58 - Depth Mapping with Depth Anything v2 How depth mapping is integrated using control net auxiliary nodes to inform the warp perspective. 5:37 - Setting Keyframes for Trajectories Practical guide on setting specific keyframes (e.g., 25th, 49th, 73rd) to define how the camera moves through space. 7:32 - Video-to-Video Transformation Showcase Applying these camera modifications to existing videos, such as character shots and futuristic environments. 8:36 - Multi-Keyframe Precision &amp; JSON Data Advanced tips on right-clicking to add points, adjusting linear vs. smooth interpolation, and manually editing JSON data for frame matching. 10:40 - Final Upscaling &amp; Side-by-Side Comparison Review of the final output after two-pass sampling (upscaling to 1080p) and comparing the original static shots to the new dynamic results. --------------------------------------------------------------------------------------------------------------- Local Workstation GPU : https:\/\/amzn.to\/3XfXsAO --------------------------------------------------------------------------------------------------------------- If You Like tutorial like this, You Can Support Our Work In Patreon: https:\/\/www.patreon.com\/c\/aifuturetech\"><\/div><\/div><div class=\"lL\" style=\"max-width:100%;width:853px;margin:5px auto;\"><\/div><figcaption><\/figcaption><\/figure>\n\n<figure class=\"wp-block-embed-youtube wp-block-embed is-type-video is-provider-youtube wp-embed-aspect-16-9 wp-has-aspect-ratio\"><div class=\"lyte-wrapper\" title=\"Generate multiple camera angles from a single Clip - on 8GB VRAM\" style=\"width:853px;max-width:100%;margin:5px auto;\"><div class=\"lyMe\" id=\"WYL_4JCLPOJTQp8\" itemprop=\"video\" itemscope itemtype=\"https:\/\/schema.org\/VideoObject\"><div><meta itemprop=\"thumbnailUrl\" content=\"https:\/\/infernews.com\/blog\/wp-content\/plugins\/wp-youtube-lyte\/lyteCache.php?origThumbUrl=https%3A%2F%2Fi.ytimg.com%2Fvi%2F4JCLPOJTQp8%2Fhqdefault.jpg\" \/><meta itemprop=\"embedURL\" content=\"https:\/\/www.youtube.com\/embed\/4JCLPOJTQp8\" \/><meta itemprop=\"duration\" content=\"PT8M21S\" \/><meta itemprop=\"uploadDate\" content=\"2026-07-18T12:30:35Z\" \/><\/div><div id=\"lyte_4JCLPOJTQp8\" data-src=\"https:\/\/infernews.com\/blog\/wp-content\/plugins\/wp-youtube-lyte\/lyteCache.php?origThumbUrl=https%3A%2F%2Fi.ytimg.com%2Fvi%2F4JCLPOJTQp8%2Fhqdefault.jpg\" class=\"pL\"><div class=\"tC\"><div class=\"tT\" itemprop=\"name\">Generate multiple camera angles from a single Clip - on 8GB VRAM<\/div><\/div><button tabindex=\"0\" class=\"play\"><\/button><div class=\"ctrl\"><div class=\"Lctrl\"><\/div><div class=\"Rctrl\"><\/div><\/div><\/div><noscript><a href=\"https:\/\/youtu.be\/4JCLPOJTQp8\" rel=\"nofollow\"><img loading=\"lazy\" decoding=\"async\" src=\"https:\/\/infernews.com\/blog\/wp-content\/plugins\/wp-youtube-lyte\/lyteCache.php?origThumbUrl=https%3A%2F%2Fi.ytimg.com%2Fvi%2F4JCLPOJTQp8%2F0.jpg\" alt=\"Generate multiple camera angles from a single Clip - on 8GB VRAM\" width=\"853\" height=\"460\" \/><br \/>Watch this video on YouTube<\/a><\/noscript><meta itemprop=\"description\" content=\"A tutorial of LTX-2.3, Cross\u2011View Prompt LoRA allows you to reposition the camera within a video clip. LoRA: https:\/\/huggingface.co\/Cseti\/LTX2.3-22B_IC-LoRA-CrossView-Prompt Workflow: https:\/\/huggingface.co\/datasets\/Cseti\/ComfyUI-Workflows\/tree\/main\/ltx\/2.3\/ic-lora-crossview-v1-pilot Related videos: Camera Control like a Cameraman - on 8GB VRAM https:\/\/youtu.be\/7hLttTSlnhE LTX-2.3 4x Keyframes Video on 8GB VRAM https:\/\/youtu.be\/E5TylOdwYG8 LTX-2.3 Audio mixed with TTS on 8GB VRAM https:\/\/youtu.be\/ctw9vXqDoSM Output files of the tutorial are uploaded to Patreon, free to join. https:\/\/www.patreon.com\/bigboss97 ======== Please check out my affiliate links and support this channel: https:\/\/tiny.cc\/go2shop https:\/\/tiny.cc\/go2ebay You can also buy me a coffee \u2615, thanks \ud83d\ude00! https:\/\/ko-fi.com\/bigboss97 Paypal donation http:\/\/tiny.cc\/support-bigboss97 https:\/\/www.patreon.com\/bigboss97\"><\/div><\/div><div class=\"lL\" style=\"max-width:100%;width:853px;margin:5px auto;\"><\/div><figcaption><\/figcaption><\/figure>\n\n\n<hr class=\"wp-block-separator has-alpha-channel-opacity\"\/>\n\n\n\n<h2 class=\"wp-block-heading\">LTX IC-LoRA<\/h2>\n\n\n<figure class=\"wp-block-embed-youtube wp-block-embed is-type-video is-provider-youtube wp-embed-aspect-16-9 wp-has-aspect-ratio\"><div class=\"lyte-wrapper\" title=\"LTX 2.3 IC-LoRA: Remove People &amp;amp; Objects from Videos! (Free 8GB VRAM ComfyUI Workflow)\" style=\"width:853px;max-width:100%;margin:5px auto;\"><div class=\"lyMe\" id=\"WYL_MJXnn4xT8mc\" itemprop=\"video\" itemscope itemtype=\"https:\/\/schema.org\/VideoObject\"><div><meta itemprop=\"thumbnailUrl\" content=\"https:\/\/infernews.com\/blog\/wp-content\/plugins\/wp-youtube-lyte\/lyteCache.php?origThumbUrl=https%3A%2F%2Fi.ytimg.com%2Fvi%2FMJXnn4xT8mc%2Fhqdefault.jpg\" \/><meta itemprop=\"embedURL\" content=\"https:\/\/www.youtube.com\/embed\/MJXnn4xT8mc\" \/><meta itemprop=\"duration\" content=\"PT8M32S\" \/><meta itemprop=\"uploadDate\" content=\"2026-07-26T14:57:55Z\" \/><\/div><meta itemprop=\"accessibilityFeature\" content=\"captions\" \/><div id=\"lyte_MJXnn4xT8mc\" data-src=\"https:\/\/infernews.com\/blog\/wp-content\/plugins\/wp-youtube-lyte\/lyteCache.php?origThumbUrl=https%3A%2F%2Fi.ytimg.com%2Fvi%2FMJXnn4xT8mc%2Fhqdefault.jpg\" class=\"pL\"><div class=\"tC\"><div class=\"tT\" itemprop=\"name\">LTX 2.3 IC-LoRA: Remove People &amp; Objects from Videos! (Free 8GB VRAM ComfyUI Workflow)<\/div><\/div><button tabindex=\"0\" class=\"play\"><\/button><div class=\"ctrl\"><div class=\"Lctrl\"><\/div><div class=\"Rctrl\"><\/div><\/div><\/div><noscript><a href=\"https:\/\/youtu.be\/MJXnn4xT8mc\" rel=\"nofollow\"><img loading=\"lazy\" decoding=\"async\" src=\"https:\/\/infernews.com\/blog\/wp-content\/plugins\/wp-youtube-lyte\/lyteCache.php?origThumbUrl=https%3A%2F%2Fi.ytimg.com%2Fvi%2FMJXnn4xT8mc%2F0.jpg\" alt=\"LTX 2.3 IC-LoRA: Remove People &amp;amp; Objects from Videos! (Free 8GB VRAM ComfyUI Workflow)\" width=\"853\" height=\"460\" \/><br \/>Watch this video on YouTube<\/a><\/noscript><meta itemprop=\"description\" content=\"\ud83d\udcc2 DOWNLOAD the 8GB VRAM optimized LTX Director 2.0 IC-LoRA ComfyUI Workflow: https:\/\/drive.google.com\/file\/d\/1q9lX07l4d5XIFc6CO17R3zZn7fyIKVdm\/view?usp=drive_link Official LTX 2.3 Clean Plate IC-LoRA: https:\/\/huggingface.co\/Lightricks\/LTX-2.3-22b-IC-LoRA-Clean-Plate Remove people and unwanted objects from videos locally with the new LTX 2.3 Clean Plate IC-LoRA in ComfyUI. This free LTX Director workflow was tested on an RTX 5060 laptop with only 8GB VRAM. Designed for video-to-video AI editing and VFX clean-plate generation, this IC-LoRA uses the original clip as its reference and reconstructs the scene without people or other dynamic subjects. There is no manual masking, tracking, rotoscoping, or frame-by-frame cleanup. In this complete tutorial, I test the workflow across crowded city streets, moving-camera footage, a rooftop apiary, boat workshop, mangrove boardwalk, control room, bakery, living room, road-maintenance scene, solar farm, and horse stable. The demonstrations include people, vehicles, equipment, furniture, animals, foreground obstructions, repeating structures, reflections, and hidden background areas. The video covers: * Installing the LTX 2.3 Clean Plate IC-LoRA * Loading source videos through LTX Director * The exact positive prompt used with the distilled workflow * LoRA strength behavior at 1.0 and 1.5 * Practical 720p, 25 FPS settings for 8GB VRAM * Removing people and dynamic objects without a mask * Working with static and controlled moving-camera shots * Choosing suitable subject sizes and compositions * Recognizing ghosted remnants and reconstructed details * The realistic strengths and limitations of full-frame AI video reconstruction LTX regenerates the complete frame instead of preserving every original pixel. The resulting background can look coherent and believable, but smaller details\u2014including text, reflections, ropes, furniture, food, screens, and hidden structures\u2014may change. This tutorial shows the successful results alongside the limitations that matter for AI video editing, clean plates, compositing, post-production, and VFX cleanup. Tensor Alchemist is building toward 10,000 subscribers with practical ComfyUI and local AI tutorials tested on real consumer hardware. If this workflow helps your projects, your subscription supports more honest low-VRAM testing and free workflow releases. \ud83d\udccc Timestamps: 0:00 - IC LoRA Clean Plate Previews 2:39 - Clean Plate IC LoRA Breakdown 3:49 - DIrector 2.0 IC-LoRA Workflow 5:45 - Prompting 6:08 - LoRA Strength 8:09 - Achieving The Best results \ud83d\udcbb My Setup : Laptop: Gigabyte A16 GPU: NVIDIA RTX 5060 (8GB VRAM) RAM: DDR5 32GB (Dual Channel) \ud83d\udcbc Business Inquiries &amp; Sponsorships For brand partnerships, hardware evaluations, software reviews, or enterprise tool placements targeting an engaged audience of local machine learning developers, AI researchers, and digital media professionals: please contact me via email. #ltx23 #comfyui #aivdeo #ltxvideo #aivideoediting #iclora #8gbvram #rtx5060\"><\/div><\/div><div class=\"lL\" style=\"max-width:100%;width:853px;margin:5px auto;\"><\/div><figcaption><\/figcaption><\/figure>\n\n<figure class=\"wp-block-embed-youtube wp-block-embed is-type-video is-provider-youtube wp-embed-aspect-16-9 wp-has-aspect-ratio\"><div class=\"lyte-wrapper\" title=\"I Tested the New LTX 2.3 Ingredients LoRA &mdash; This Is Huge for AI Filmmaking\" style=\"width:853px;max-width:100%;margin:5px auto;\"><div class=\"lyMe\" id=\"WYL_guxp6uFqwn8\" itemprop=\"video\" itemscope itemtype=\"https:\/\/schema.org\/VideoObject\"><div><meta itemprop=\"thumbnailUrl\" content=\"https:\/\/infernews.com\/blog\/wp-content\/plugins\/wp-youtube-lyte\/lyteCache.php?origThumbUrl=https%3A%2F%2Fi.ytimg.com%2Fvi%2Fguxp6uFqwn8%2Fhqdefault.jpg\" \/><meta itemprop=\"embedURL\" content=\"https:\/\/www.youtube.com\/embed\/guxp6uFqwn8\" \/><meta itemprop=\"duration\" content=\"PT12M21S\" \/><meta itemprop=\"uploadDate\" content=\"2026-07-02T13:02:30Z\" \/><\/div><div id=\"lyte_guxp6uFqwn8\" data-src=\"https:\/\/infernews.com\/blog\/wp-content\/plugins\/wp-youtube-lyte\/lyteCache.php?origThumbUrl=https%3A%2F%2Fi.ytimg.com%2Fvi%2Fguxp6uFqwn8%2Fhqdefault.jpg\" class=\"pL\"><div class=\"tC\"><div class=\"tT\" itemprop=\"name\">I Tested the New LTX 2.3 Ingredients LoRA \u2014 This Is Huge for AI Filmmaking<\/div><\/div><button tabindex=\"0\" class=\"play\"><\/button><div class=\"ctrl\"><div class=\"Lctrl\"><\/div><div class=\"Rctrl\"><\/div><\/div><\/div><noscript><a href=\"https:\/\/youtu.be\/guxp6uFqwn8\" rel=\"nofollow\"><img loading=\"lazy\" decoding=\"async\" src=\"https:\/\/infernews.com\/blog\/wp-content\/plugins\/wp-youtube-lyte\/lyteCache.php?origThumbUrl=https%3A%2F%2Fi.ytimg.com%2Fvi%2Fguxp6uFqwn8%2F0.jpg\" alt=\"I Tested the New LTX 2.3 Ingredients LoRA &mdash; This Is Huge for AI Filmmaking\" width=\"853\" height=\"460\" \/><br \/>Watch this video on YouTube<\/a><\/noscript><meta itemprop=\"description\" content=\"In this video, I show you the new LTX 2.3 Ingredients IC-LoRA \u2014 one of the most interesting tools for creating more consistent AI video scenes inside ComfyUI. This LoRA is not just another style LoRA. It works with a reference sheet, where you can prepare all the important \u201cingredients\u201d of your scene: characters, outfits, props, objects, locations, colors, lighting, and visual style. Then LTX 2.3 uses those references to generate a new cinematic video while trying to preserve consistency across the entire shot. This is a big step forward for AI filmmaking, because one of the hardest problems in AI video is keeping the same characters, faces, clothes, props, and environments stable from frame to frame. In this tutorial, I show: How the LTX 2.3 Ingredients LoRA works How to prepare a proper reference sheet How to use characters, props, outfits, and locations as scene ingredients How to connect the workflow in ComfyUI How to generate a cinematic AI video scene Why this can be useful for animated series, short films, product videos, and local AI video production You can download the workflow, model links, and reference images from my Patreon post here: https:\/\/www.patreon.com\/AntiMatterComfy\/posts\/ltx-2-3-lora-in-162691572?utm_medium=clipboard_copy&amp;utm_source=copyLink&amp;utm_campaign=postshare_creator&amp;utm_content=join_link This is a free workflow post, so you can test it locally and experiment with your own characters and scenes. If you want more advanced system prompts, agent instructions, and professional prompt setups for generating better Ingredients LoRA reference sheets and video prompts, check out my paid Patreon tier. Thanks for watching, and subscribe for more ComfyUI, LTX 2.3, AI video, and local AI filmmaking workflows\"><\/div><\/div><div class=\"lL\" style=\"max-width:100%;width:853px;margin:5px auto;\"><\/div><figcaption><\/figcaption><\/figure>\n\n<figure class=\"wp-block-embed-youtube wp-block-embed is-type-video is-provider-youtube wp-embed-aspect-16-9 wp-has-aspect-ratio\"><div class=\"lyte-wrapper\" title=\"LTX 2.3 New Camera Control Using IC-LoRA + Krea 2 + Boogu Edit Mix Models Play Time!\" style=\"width:853px;max-width:100%;margin:5px auto;\"><div class=\"lyMe\" id=\"WYL_Lp9-o3cEVRQ\" itemprop=\"video\" itemscope itemtype=\"https:\/\/schema.org\/VideoObject\"><div><meta itemprop=\"thumbnailUrl\" content=\"https:\/\/infernews.com\/blog\/wp-content\/plugins\/wp-youtube-lyte\/lyteCache.php?origThumbUrl=https%3A%2F%2Fi.ytimg.com%2Fvi%2FLp9-o3cEVRQ%2Fhqdefault.jpg\" \/><meta itemprop=\"embedURL\" content=\"https:\/\/www.youtube.com\/embed\/Lp9-o3cEVRQ\" \/><meta itemprop=\"duration\" content=\"PT8M16S\" \/><meta itemprop=\"uploadDate\" content=\"2026-06-28T13:00:19Z\" \/><\/div><div id=\"lyte_Lp9-o3cEVRQ\" data-src=\"https:\/\/infernews.com\/blog\/wp-content\/plugins\/wp-youtube-lyte\/lyteCache.php?origThumbUrl=https%3A%2F%2Fi.ytimg.com%2Fvi%2FLp9-o3cEVRQ%2Fhqdefault.jpg\" class=\"pL\"><div class=\"tC\"><div class=\"tT\" itemprop=\"name\">LTX 2.3 New Camera Control Using IC-LoRA + Krea 2 + Boogu Edit Mix Models Play Time!<\/div><\/div><button tabindex=\"0\" class=\"play\"><\/button><div class=\"ctrl\"><div class=\"Lctrl\"><\/div><div class=\"Rctrl\"><\/div><\/div><\/div><noscript><a href=\"https:\/\/youtu.be\/Lp9-o3cEVRQ\" rel=\"nofollow\"><img loading=\"lazy\" decoding=\"async\" src=\"https:\/\/infernews.com\/blog\/wp-content\/plugins\/wp-youtube-lyte\/lyteCache.php?origThumbUrl=https%3A%2F%2Fi.ytimg.com%2Fvi%2FLp9-o3cEVRQ%2F0.jpg\" alt=\"LTX 2.3 New Camera Control Using IC-LoRA + Krea 2 + Boogu Edit Mix Models Play Time!\" width=\"853\" height=\"460\" \/><br \/>Watch this video on YouTube<\/a><\/noscript><meta itemprop=\"description\" content=\"This video walks through how to use the IC-LoRA Cameraman V2 from Hugging Face to replicate real camera movements from reference videos and apply them to AI-generated scenes in ComfyUI. The workflow combines LTX Video 2.3 22B with Krea 2 for image generation and Boogu Image 0.1 Edit for scene compositing, creating dynamic camera shots without writing lengthy text prompts. This tutorial is for AI video creators and ComfyUI users who want more control over camera motion in their generated videos. If you&#039;ve been relying purely on text prompts to describe camera movements and finding the results inconsistent, this video-to-video approach gives you exact control using real footage as your camera reference. Text-to-video camera control has been one of the weakest points in AI video generation. Describing complex camera pans, tracking shots, and dynamic angles in prompts rarely produces what you actually want. The IC-LoRA Cameraman V2 solves this by letting you extract camera motion from any video and transfer it to your generated scenes, giving you broadcast-quality camera work with minimal effort. Resources: Krea 2 https:\/\/www.youtube.com\/watch?v=k8-9qGbPfpM Boogu-Image-0.1 Edit https:\/\/www.youtube.com\/watch?v=m4fK-IAoFd0 LTX-Video 2.3 22B \u2014 IC-LoRA: Cameraman v2 https:\/\/huggingface.co\/Cseti\/LTX2.3-22B_IC-LoRA-Cameraman_v2 Attached example freebie workflows. https:\/\/www.patreon.com\/aifuturetech\/posts\/162302034?utm_source=youtube&amp;utm_medium=video&amp;utm_campaign=20260628 -------------------------------------------------------------------------------------- Local Workstation GPU : https:\/\/amzn.to\/3XfXsAO -------------------------------------------------------------------------------------- If You Like tutorial like this, You Can Support Our Work In Patreon: https:\/\/www.patreon.com\/c\/aifuturetech\"><\/div><\/div><div class=\"lL\" style=\"max-width:100%;width:853px;margin:5px auto;\"><\/div><figcaption><\/figcaption><\/figure>\n\n<figure class=\"wp-block-embed-youtube wp-block-embed is-type-video is-provider-youtube wp-embed-aspect-16-9 wp-has-aspect-ratio\"><div class=\"lyte-wrapper\" title=\"LTX 2.3 Ingredients IC-LoRA &mdash; The NEW Reference-to-Video Model\" style=\"width:853px;max-width:100%;margin:5px auto;\"><div class=\"lyMe\" id=\"WYL_P-nRo5muRCM\" itemprop=\"video\" itemscope itemtype=\"https:\/\/schema.org\/VideoObject\"><div><meta itemprop=\"thumbnailUrl\" content=\"https:\/\/infernews.com\/blog\/wp-content\/plugins\/wp-youtube-lyte\/lyteCache.php?origThumbUrl=https%3A%2F%2Fi.ytimg.com%2Fvi%2FP-nRo5muRCM%2Fhqdefault.jpg\" \/><meta itemprop=\"embedURL\" content=\"https:\/\/www.youtube.com\/embed\/P-nRo5muRCM\" \/><meta itemprop=\"duration\" content=\"PT9M42S\" \/><meta itemprop=\"uploadDate\" content=\"2026-06-18T13:57:26Z\" \/><\/div><div id=\"lyte_P-nRo5muRCM\" data-src=\"https:\/\/infernews.com\/blog\/wp-content\/plugins\/wp-youtube-lyte\/lyteCache.php?origThumbUrl=https%3A%2F%2Fi.ytimg.com%2Fvi%2FP-nRo5muRCM%2Fhqdefault.jpg\" class=\"pL\"><div class=\"tC\"><div class=\"tT\" itemprop=\"name\">LTX 2.3 Ingredients IC-LoRA \u2014 The NEW Reference-to-Video Model<\/div><\/div><button tabindex=\"0\" class=\"play\"><\/button><div class=\"ctrl\"><div class=\"Lctrl\"><\/div><div class=\"Rctrl\"><\/div><\/div><\/div><noscript><a href=\"https:\/\/youtu.be\/P-nRo5muRCM\" rel=\"nofollow\"><img loading=\"lazy\" decoding=\"async\" src=\"https:\/\/infernews.com\/blog\/wp-content\/plugins\/wp-youtube-lyte\/lyteCache.php?origThumbUrl=https%3A%2F%2Fi.ytimg.com%2Fvi%2FP-nRo5muRCM%2F0.jpg\" alt=\"LTX 2.3 Ingredients IC-LoRA &mdash; The NEW Reference-to-Video Model\" width=\"853\" height=\"460\" \/><br \/>Watch this video on YouTube<\/a><\/noscript><meta itemprop=\"description\" content=\"Lightricks just dropped a major update to the LTX 2.3 ecosystem: a completely rebuilt LoRA trainer with native IC-LoRA support, and a brand-new &quot;Ingredients&quot; model for reference-to-video generation. In this video, we walk through the full new trainer pipeline, show you how to use the Ingredients IC-LoRA to generate multi-character video from a single reference sheet, and compare it head-to-head with Lykon Studio&#039;s MSR-LoRA. We also cover the practical ComfyUI workflow, why Prompt Relay doesn&#039;t work with Ingredients right now, and how to set up your own reference sheet using Ideogram 4 and AI-assisted prompt generation. This content is built for AI video creators, ComfyUI users, and anyone working with the LTX model family. Whether you&#039;re generating multi-character scenes, training your own video LoRAs, or just trying to figure out which reference-to-video approach is best for your workflow, this video gives you a hands-on breakdown with real examples. No fluff \u2014 just what works, what doesn&#039;t, and how to get the best results on consumer hardware. This matters because the new LTX trainer opens the door for anyone to train custom video LoRAs, not just rely on community releases. With built-in agent instructions for AI-assisted training, support for audio and video conditioning, and a growing Creative Lab collection, LTX 2.3 is becoming a serious production pipeline. Understanding the Ingredients model now \u2014 and its strengths and limitations compared to alternatives like MSR-LoRA \u2014 will help you pick the right tool for your projects and prepare for the next wave of custom-trained models. News About New LTX Trainer https:\/\/x.com\/ltx_io\/status\/2067284780081905685 Github https:\/\/github.com\/Lightricks\/LTX-2 Model https:\/\/huggingface.co\/Lightricks\/LTX-2.3 LoRAs and IC-LoRAs, trained on the LTX-2.3 model https:\/\/huggingface.co\/collections\/Lightricks\/ltx-23-creative-lab Lightricks\/LTX-2.3-22b-IC-LoRA-Ingredients https:\/\/huggingface.co\/Lightricks\/LTX-2.3-22b-IC-LoRA-Ingredients Mod Workflow based on Lightricks Github workflow https:\/\/www.patreon.com\/aifuturetech\/posts\/ltx-2-3-ic-lora-161431864 Local Workstation GPU : https:\/\/amzn.to\/3XfXsAO -------------------------------------------------------------------------------------------------------------------------------- If You Like tutorial like this, You Can Support Our Work In Patreon: https:\/\/www.patreon.com\/c\/aifuturetech\"><\/div><\/div><div class=\"lL\" style=\"max-width:100%;width:853px;margin:5px auto;\"><\/div><figcaption><\/figcaption><\/figure>\n\n\n<hr class=\"wp-block-separator has-alpha-channel-opacity\"\/>\n\n\n\n<h2 class=\"wp-block-heading\">LTX 2.3<\/h2>\n\n\n<figure class=\"wp-block-embed-youtube wp-block-embed is-type-video is-provider-youtube wp-embed-aspect-16-9 wp-has-aspect-ratio\"><div class=\"lyte-wrapper\" title=\"LTX2.3\u4e00\u952e\u6574\u5408\u5305\uff0c\u6570\u5b57\u4eba\u5bf9\u53e3\u578b\uff0cAI\u5531\u6b4c\uff0c\u9996\u5c3e\u5e27\uff0c\u56fe\u751f\u89c6\u9891\uff0c\u6587\u751f\u89c6\u9891\uff0c\u56fe\u7247\u7f16\u8f91\uff0c\u6587\u751f\u56fe\uff0c\u652f\u630150\u7cfb\u663e\u5361\uff0c8G\u663e\u5b58\u53ef\u7528\uff0c64G\u5185\u5b58\u3002\" style=\"width:853px;max-width:100%;margin:5px auto;\"><div class=\"lyMe\" id=\"WYL_aGHew8uDbZs\" itemprop=\"video\" itemscope itemtype=\"https:\/\/schema.org\/VideoObject\"><div><meta itemprop=\"thumbnailUrl\" content=\"https:\/\/infernews.com\/blog\/wp-content\/plugins\/wp-youtube-lyte\/lyteCache.php?origThumbUrl=https%3A%2F%2Fi.ytimg.com%2Fvi%2FaGHew8uDbZs%2Fhqdefault.jpg\" \/><meta itemprop=\"embedURL\" content=\"https:\/\/www.youtube.com\/embed\/aGHew8uDbZs\" \/><meta itemprop=\"duration\" content=\"PT6M9S\" \/><meta itemprop=\"uploadDate\" content=\"2026-06-13T15:32:59Z\" \/><\/div><meta itemprop=\"accessibilityFeature\" content=\"captions\" \/><div id=\"lyte_aGHew8uDbZs\" data-src=\"https:\/\/infernews.com\/blog\/wp-content\/plugins\/wp-youtube-lyte\/lyteCache.php?origThumbUrl=https%3A%2F%2Fi.ytimg.com%2Fvi%2FaGHew8uDbZs%2Fhqdefault.jpg\" class=\"pL\"><div class=\"tC\"><div class=\"tT\" itemprop=\"name\">LTX2.3\u4e00\u952e\u6574\u5408\u5305\uff0c\u6570\u5b57\u4eba\u5bf9\u53e3\u578b\uff0cAI\u5531\u6b4c\uff0c\u9996\u5c3e\u5e27\uff0c\u56fe\u751f\u89c6\u9891\uff0c\u6587\u751f\u89c6\u9891\uff0c\u56fe\u7247\u7f16\u8f91\uff0c\u6587\u751f\u56fe\uff0c\u652f\u630150\u7cfb\u663e\u5361\uff0c8G\u663e\u5b58\u53ef\u7528\uff0c64G\u5185\u5b58\u3002<\/div><\/div><button tabindex=\"0\" class=\"play\"><\/button><div class=\"ctrl\"><div class=\"Lctrl\"><\/div><div class=\"Rctrl\"><\/div><\/div><\/div><noscript><a href=\"https:\/\/youtu.be\/aGHew8uDbZs\" rel=\"nofollow\"><img loading=\"lazy\" decoding=\"async\" src=\"https:\/\/infernews.com\/blog\/wp-content\/plugins\/wp-youtube-lyte\/lyteCache.php?origThumbUrl=https%3A%2F%2Fi.ytimg.com%2Fvi%2FaGHew8uDbZs%2F0.jpg\" alt=\"LTX2.3\u4e00\u952e\u6574\u5408\u5305\uff0c\u6570\u5b57\u4eba\u5bf9\u53e3\u578b\uff0cAI\u5531\u6b4c\uff0c\u9996\u5c3e\u5e27\uff0c\u56fe\u751f\u89c6\u9891\uff0c\u6587\u751f\u89c6\u9891\uff0c\u56fe\u7247\u7f16\u8f91\uff0c\u6587\u751f\u56fe\uff0c\u652f\u630150\u7cfb\u663e\u5361\uff0c8G\u663e\u5b58\u53ef\u7528\uff0c64G\u5185\u5b58\u3002\" width=\"853\" height=\"460\" \/><br \/>Watch this video on YouTube<\/a><\/noscript><meta itemprop=\"description\" content=\"1\u3001\u6548\u679c\u975e\u5e38\u60ca\u8273 LTX2.3\uff1a\u6570\u5b57\u4eba\uff0c\u56fe\u7247\u3001\u6587\u5b57\u751f\u6210\u89c6\u9891\uff0c\u6574\u5408\u5305 + \u4e91\u7aef\u6211\u90fd\u505a\u597d\u4e86\uff01 2\u3001\u4e91\u7aef\u4f53\u9a8c\uff1ahttps:\/\/www.compshare.cn\/images\/gbEUSxYJatWb?referral_code=bZRguxQSRDBJHgp5vAE7u&amp;ytag=GPU_YY_YX_bl_yuziyuetalk0614 3\u3001\u6574\u5408\u5305\uff1a\u5173\u6ce8 + \u4e09\u8fde \u53d1\u9001\u79c1\u4fe1\u5173\u952e\u5b57 \u201cltx23\u201d \u81ea\u53d6 \u65f6\u95f4\u8f74\uff1a 00:00 \u5f00\u573a 01:44 \u6574\u5408\u5305 04:49 \u663e\u5b58\u5360\u7528 05:09 \u4e91\u7aef\u4f7f\u7528\"><\/div><\/div><div class=\"lL\" style=\"max-width:100%;width:853px;margin:5px auto;\"><\/div><figcaption><\/figcaption><\/figure>\n\n<figure class=\"wp-block-embed-youtube wp-block-embed is-type-video is-provider-youtube wp-embed-aspect-16-9 wp-has-aspect-ratio\"><div class=\"lyte-wrapper\" title=\"LTX 2.3 Character Motion Transfer | Video to Image Animation\" style=\"width:853px;max-width:100%;margin:5px auto;\"><div class=\"lyMe\" id=\"WYL_JiZ6GX_HwPg\" itemprop=\"video\" itemscope itemtype=\"https:\/\/schema.org\/VideoObject\"><div><meta itemprop=\"thumbnailUrl\" content=\"https:\/\/infernews.com\/blog\/wp-content\/plugins\/wp-youtube-lyte\/lyteCache.php?origThumbUrl=https%3A%2F%2Fi.ytimg.com%2Fvi%2FJiZ6GX_HwPg%2Fhqdefault.jpg\" \/><meta itemprop=\"embedURL\" content=\"https:\/\/www.youtube.com\/embed\/JiZ6GX_HwPg\" \/><meta itemprop=\"duration\" content=\"PT10M30S\" \/><meta itemprop=\"uploadDate\" content=\"2026-06-12T12:14:16Z\" \/><\/div><meta itemprop=\"accessibilityFeature\" content=\"captions\" \/><div id=\"lyte_JiZ6GX_HwPg\" data-src=\"https:\/\/infernews.com\/blog\/wp-content\/plugins\/wp-youtube-lyte\/lyteCache.php?origThumbUrl=https%3A%2F%2Fi.ytimg.com%2Fvi%2FJiZ6GX_HwPg%2Fhqdefault.jpg\" class=\"pL\"><div class=\"tC\"><div class=\"tT\" itemprop=\"name\">LTX 2.3 Character Motion Transfer | Video to Image Animation<\/div><\/div><button tabindex=\"0\" class=\"play\"><\/button><div class=\"ctrl\"><div class=\"Lctrl\"><\/div><div class=\"Rctrl\"><\/div><\/div><\/div><noscript><a href=\"https:\/\/youtu.be\/JiZ6GX_HwPg\" rel=\"nofollow\"><img loading=\"lazy\" decoding=\"async\" src=\"https:\/\/infernews.com\/blog\/wp-content\/plugins\/wp-youtube-lyte\/lyteCache.php?origThumbUrl=https%3A%2F%2Fi.ytimg.com%2Fvi%2FJiZ6GX_HwPg%2F0.jpg\" alt=\"LTX 2.3 Character Motion Transfer | Video to Image Animation\" width=\"853\" height=\"460\" \/><br \/>Watch this video on YouTube<\/a><\/noscript><meta itemprop=\"description\" content=\"\ud83c\udf1fLTX Prompt Timeline: Turn 5s Clips into Full Videos\ud83c\udf1f In this video, I demonstrate a character motion transfer workflow using LTX 2.3 and a LoRA. The workflow takes: A reference image of a character A driving video and transfers the motion from the video to the character in the image, creating an animated result. I also explain the workflow and share the files used so you can try it yourself. #LTX23 #LTXVideo #AIAnimation #MotionTransfer #ComfyUI #LoRA #AIVideo #CharacterAnimation #OpenSourceAI #GenerativeAI \ud83d\udc49 I\u2019ve shared two examples to demonstrate how scene changes, camera motion, and continuity can be controlled using this approach. \ud83d\ude80 Resources: Workflow, image and video I have used. https:\/\/discord.com\/channels\/1414348074477949022\/1514957453249347716 \u2728 https:\/\/huggingface.co\/datasets\/Yogesh-DevHub\/LTX2.3\/tree\/main\/LTX-IV2V-Transfer-Motion Install Sage Attention Video - 1: https:\/\/youtu.be\/-S39owjSsMo Video - 2: https:\/\/youtu.be\/b43GLxkbg6o Stay tuned for the next update \ud83d\udc40 #comfyui #aivideo #aiimage #aitools \ud83d\udcbb Hardware Used: (i5 14400F, 64GB RAM, RTX 5060 TI 16GB) \ud83d\udce5 Get Started with Comfy UI: https:\/\/youtu.be\/grzK5mBitzs \ud83d\udcac Drop a comment below! \ud83d\udd14 Stay Updated: Stay connected to find the latest AI applications! Happy creating! \ud83c\udfa8\u2728 Playlist For ComfyUI - Stable Diffusion https:\/\/www.youtube.com\/playlist?list=PLPFN04WspxqvpVeMmJ1vHqG-uf4C4dIb0 Playlist For Krita AI Diffusion https:\/\/www.youtube.com\/playlist?list=PLPFN04WspxqvFhJDIXvIDZ3yveShMvgss Playlist For Fooocus AI - Stable Diffusion https:\/\/www.youtube.com\/playlist?list=PLPFN04WspxqsslRSpiLmwGR8QTpDYNv7z Playlist For Forge UI https:\/\/www.youtube.com\/playlist?list=PLPFN04WspxqvpZmxPPjkxV-BkuOy2FF2i\"><\/div><\/div><div class=\"lL\" style=\"max-width:100%;width:853px;margin:5px auto;\"><\/div><figcaption><\/figcaption><\/figure>\n\n<figure class=\"wp-block-embed-youtube wp-block-embed is-type-video is-provider-youtube wp-embed-aspect-16-9 wp-has-aspect-ratio\"><div class=\"lyte-wrapper\" title=\"BERNINI ! Video + Image Ref 2 Video editor in ComyUI (DEEP DIVE)\" style=\"width:853px;max-width:100%;margin:5px auto;\"><div class=\"lyMe\" id=\"WYL_5oQ4F4G9MaE\" itemprop=\"video\" itemscope itemtype=\"https:\/\/schema.org\/VideoObject\"><div><meta itemprop=\"thumbnailUrl\" content=\"https:\/\/infernews.com\/blog\/wp-content\/plugins\/wp-youtube-lyte\/lyteCache.php?origThumbUrl=https%3A%2F%2Fi.ytimg.com%2Fvi%2F5oQ4F4G9MaE%2Fhqdefault.jpg\" \/><meta itemprop=\"embedURL\" content=\"https:\/\/www.youtube.com\/embed\/5oQ4F4G9MaE\" \/><meta itemprop=\"duration\" content=\"PT17M14S\" \/><meta itemprop=\"uploadDate\" content=\"2026-06-07T15:56:54Z\" \/><\/div><div id=\"lyte_5oQ4F4G9MaE\" data-src=\"https:\/\/infernews.com\/blog\/wp-content\/plugins\/wp-youtube-lyte\/lyteCache.php?origThumbUrl=https%3A%2F%2Fi.ytimg.com%2Fvi%2F5oQ4F4G9MaE%2Fhqdefault.jpg\" class=\"pL\"><div class=\"tC\"><div class=\"tT\" itemprop=\"name\">BERNINI ! Video + Image Ref 2 Video editor in ComyUI (DEEP DIVE)<\/div><\/div><button tabindex=\"0\" class=\"play\"><\/button><div class=\"ctrl\"><div class=\"Lctrl\"><\/div><div class=\"Rctrl\"><\/div><\/div><\/div><noscript><a href=\"https:\/\/youtu.be\/5oQ4F4G9MaE\" rel=\"nofollow\"><img loading=\"lazy\" decoding=\"async\" src=\"https:\/\/infernews.com\/blog\/wp-content\/plugins\/wp-youtube-lyte\/lyteCache.php?origThumbUrl=https%3A%2F%2Fi.ytimg.com%2Fvi%2F5oQ4F4G9MaE%2F0.jpg\" alt=\"BERNINI ! Video + Image Ref 2 Video editor in ComyUI (DEEP DIVE)\" width=\"853\" height=\"460\" \/><br \/>Watch this video on YouTube<\/a><\/noscript><meta itemprop=\"description\" content=\"BRILLIANT ! Bernini can edit an existing video by using an image reference to change the video. Here we look at Bernini step by step with real results If you liked this video, please consider clicking the LIKE button and subscribing, thanks. LEAVE A COMMENT. Join our &quot;Get Going Fast&quot; Discord server here for the workflow and chat https:\/\/discord.gg\/nn3RRcVy5c For one-click installs to Get Going Fast visit The AI Hobby Guy here and create a FREE account https:\/\/getgoingfast.pro\/ #bernini #wan22 #zimage #firered #comfyui #bernini\"><\/div><\/div><div class=\"lL\" style=\"max-width:100%;width:853px;margin:5px auto;\"><\/div><figcaption><\/figcaption><\/figure>\n\n<figure class=\"wp-block-embed-youtube wp-block-embed is-type-video is-provider-youtube wp-embed-aspect-16-9 wp-has-aspect-ratio\"><div class=\"lyte-wrapper\" title=\"How to Use Exciting LTX 2.3 LoRAs for Stunning AI Videos\" style=\"width:853px;max-width:100%;margin:5px auto;\"><div class=\"lyMe\" id=\"WYL_zA55q3zG930\" itemprop=\"video\" itemscope itemtype=\"https:\/\/schema.org\/VideoObject\"><div><meta itemprop=\"thumbnailUrl\" content=\"https:\/\/infernews.com\/blog\/wp-content\/plugins\/wp-youtube-lyte\/lyteCache.php?origThumbUrl=https%3A%2F%2Fi.ytimg.com%2Fvi%2FzA55q3zG930%2Fhqdefault.jpg\" \/><meta itemprop=\"embedURL\" content=\"https:\/\/www.youtube.com\/embed\/zA55q3zG930\" \/><meta itemprop=\"duration\" content=\"PT5M16S\" \/><meta itemprop=\"uploadDate\" content=\"2026-06-12T16:40:57Z\" \/><\/div><meta itemprop=\"accessibilityFeature\" content=\"captions\" \/><div id=\"lyte_zA55q3zG930\" data-src=\"https:\/\/infernews.com\/blog\/wp-content\/plugins\/wp-youtube-lyte\/lyteCache.php?origThumbUrl=https%3A%2F%2Fi.ytimg.com%2Fvi%2FzA55q3zG930%2Fhqdefault.jpg\" class=\"pL\"><div class=\"tC\"><div class=\"tT\" itemprop=\"name\">How to Use Exciting LTX 2.3 LoRAs for Stunning AI Videos<\/div><\/div><button tabindex=\"0\" class=\"play\"><\/button><div class=\"ctrl\"><div class=\"Lctrl\"><\/div><div class=\"Rctrl\"><\/div><\/div><\/div><noscript><a href=\"https:\/\/youtu.be\/zA55q3zG930\" rel=\"nofollow\"><img loading=\"lazy\" decoding=\"async\" src=\"https:\/\/infernews.com\/blog\/wp-content\/plugins\/wp-youtube-lyte\/lyteCache.php?origThumbUrl=https%3A%2F%2Fi.ytimg.com%2Fvi%2FzA55q3zG930%2F0.jpg\" alt=\"How to Use Exciting LTX 2.3 LoRAs for Stunning AI Videos\" width=\"853\" height=\"460\" \/><br \/>Watch this video on YouTube<\/a><\/noscript><meta itemprop=\"description\" content=\"\ud83d\udd25 In this video, we&#039;re exploring some of the most interesting LoRA models available for LTX 2.3 and how they can transform ordinary generations into something truly unique. I&#039;ll also show you how I use an LTX Director workflow with a low-VRAM GGUF model and OmniNFT RL-LoRA to achieve better performance, improved lip-sync, and more natural character animations. \u2728 We&#039;ll dive into Cinematic Post Apocalyptic, a LoRA that turns simple footage into gritty, photorealistic survival scenes straight out of a blockbuster movie. \u2728 Next, we&#039;ll take a look at LTX2.3-IC-LORA-Dual-Character, a powerful solution for maintaining character consistency in conversations and action scenes\u2014one of the biggest challenges in AI video generation. \u2728 Need a break from filmmaking? We&#039;ll switch gears and experiment with K-pop Dance LoRA, complete with its quirks, unexpected results, and a few lessons learned through trial and error. \u2728 We&#039;ll also explore Latin Dance LoRA, which pairs beautifully with Image-to-Video workflows and custom music tracks, allowing you to create energetic and personalized dance performances. \u2728 Finally, we&#039;ll step into the neon-lit world of Synthwave LoRA, where retro-futurism, 1980s nostalgia, and vibrant aesthetics come together to create visually stunning videos. Whether you&#039;re making cinematic stories, dance videos, or experimenting with new visual styles, I hope this video gives you fresh ideas and inspiration for your next AI project. If you found this video helpful, please consider leaving a like and subscribing to the channel. Your support means a lot and helps me continue creating more AI tutorials and creative experiments. Thanks for watching. Keep creating, keep exploring, and I&#039;ll see you in the next AI adventure! \u2728 All workflows and resources are shared for free. If you\u2019d like to support the channel, you can buy me a coffee \u2764\ufe0f: https:\/\/ko-fi.com\/fututek -------------------------------------------------- \u2728Run on Cloud GPU: https:\/\/runpod.io?ref=99cuxxw1 -------------------------------------------------- \u2728Workflow and resources: https:\/\/drive.google.com\/drive\/folders\/1yxMVGqgDcJDIqSE_X61YQOdQxHLEHMGx?usp=sharing - LTX2.3 Model for low VRAM 8~12Gb (Place in \/ComfyUI\/models\/diffusion_models) https:\/\/huggingface.co\/vantagewithai\/LTX-2.3-GGUF\/blob\/main\/dev\/ltx-2-3-22b-dev-Q4_K_M.gguf - Text Encoder (Place in \/ComfyUI\/models\/text_encoders) https:\/\/huggingface.co\/Comfy-Org\/ltx-2\/resolve\/main\/split_files\/text_encoders\/gemma_3_12B_it_fp4_mixed.safetensors https:\/\/huggingface.co\/Kijai\/LTX2.3_comfy\/blob\/main\/text_encoders\/ltx-2.3_text_projection_bf16.safetensors - VAE (Place in \/ComfyUI\/models\/vae) https:\/\/huggingface.co\/Kijai\/LTX2.3_comfy\/blob\/main\/vae\/LTX23_audio_vae_bf16.safetensors https:\/\/huggingface.co\/Kijai\/LTX2.3_comfy\/blob\/main\/vae\/LTX23_video_vae_bf16.safetensors - Spatial Upscaler (Place in \/ComfyUI\/models\/latent_upscale_models) https:\/\/huggingface.co\/Lightricks\/LTX-2.3\/resolve\/main\/ltx-2.3-spatial-upscaler-x2-1.1.safetensors - Distilled Loras (Place in \/ComfyUI\/models\/loras) https:\/\/huggingface.co\/Kijai\/LTX2.3_comfy\/blob\/main\/loras\/ltx-2.3-22b-distilled-lora-dynamic_fro09_avg_rank_105_bf16.safetensors - Enhancing Loras (Place in \/ComfyUI\/models\/loras) 1. Cinematic Post-Apocalyptic: https:\/\/civitai.com\/models\/2563394\/ltx-23-cinematic-post-apocalyptic 2. Dual Characters: https:\/\/civitai.com\/models\/2500098\/ltx23-ic-lora-dual-character?modelVersionId=2810376 3. K-Pop Dance: https:\/\/civitai.com\/models\/2487939\/ltx-23-dance 4. Latin Rhythm: https:\/\/civitai.com\/models\/2649352\/ltx-23-latin-dance 5. Synthware https:\/\/civitai.com\/models\/2551439\/ltx-23-synthwave 6. OmniNFT RL-LoRA https:\/\/huggingface.co\/Kijai\/LTX2.3_comfy\/resolve\/main\/loras\/LTX-2.3-OmniNFT-RL-Lora_bf16.safetensors\"><\/div><\/div><div class=\"lL\" style=\"max-width:100%;width:853px;margin:5px auto;\"><\/div><figcaption><\/figcaption><\/figure>\n\n<figure class=\"wp-block-embed-youtube wp-block-embed is-type-video is-provider-youtube wp-embed-aspect-16-9 wp-has-aspect-ratio\"><div class=\"lyte-wrapper\" title=\"This LTX-2.3 LoRA Removes Obstructions From Video | Wan2GP Tutorial\" style=\"width:853px;max-width:100%;margin:5px auto;\"><div class=\"lyMe\" id=\"WYL_1fOPyztY768\" itemprop=\"video\" itemscope itemtype=\"https:\/\/schema.org\/VideoObject\"><div><meta itemprop=\"thumbnailUrl\" content=\"https:\/\/infernews.com\/blog\/wp-content\/plugins\/wp-youtube-lyte\/lyteCache.php?origThumbUrl=https%3A%2F%2Fi.ytimg.com%2Fvi%2F1fOPyztY768%2Fhqdefault.jpg\" \/><meta itemprop=\"embedURL\" content=\"https:\/\/www.youtube.com\/embed\/1fOPyztY768\" \/><meta itemprop=\"duration\" content=\"PT4M16S\" \/><meta itemprop=\"uploadDate\" content=\"2026-05-30T03:17:41Z\" \/><\/div><div id=\"lyte_1fOPyztY768\" data-src=\"https:\/\/infernews.com\/blog\/wp-content\/plugins\/wp-youtube-lyte\/lyteCache.php?origThumbUrl=https%3A%2F%2Fi.ytimg.com%2Fvi%2F1fOPyztY768%2Fhqdefault.jpg\" class=\"pL\"><div class=\"tC\"><div class=\"tT\" itemprop=\"name\">This LTX-2.3 LoRA Removes Obstructions From Video | Wan2GP Tutorial<\/div><\/div><button tabindex=\"0\" class=\"play\"><\/button><div class=\"ctrl\"><div class=\"Lctrl\"><\/div><div class=\"Rctrl\"><\/div><\/div><\/div><noscript><a href=\"https:\/\/youtu.be\/1fOPyztY768\" rel=\"nofollow\"><img loading=\"lazy\" decoding=\"async\" src=\"https:\/\/infernews.com\/blog\/wp-content\/plugins\/wp-youtube-lyte\/lyteCache.php?origThumbUrl=https%3A%2F%2Fi.ytimg.com%2Fvi%2F1fOPyztY768%2F0.jpg\" alt=\"This LTX-2.3 LoRA Removes Obstructions From Video | Wan2GP Tutorial\" width=\"853\" height=\"460\" \/><br \/>Watch this video on YouTube<\/a><\/noscript><meta itemprop=\"description\" content=\"In this quick demo and tutorial, I show how to use Obscura Remova, an LTX-2.3 LoRA designed to remove obstructions, blocking objects, and unwanted foreground elements from video. We\u2019ll run the LoRA inside Wan2GP, test it on obstructed video examples, and compare the before\/after results to see how well LTX-2.3 can reconstruct a cleaner video frame. This is useful for AI video workflows where you want to remove fences, foreground objects, text, visual clutter, or other elements blocking the subject. \ud83d\udcda Resources: \ud83d\udd17 Setting Up Your Wan2GP Environment on RunPod: https:\/\/youtu.be\/_fZqf4bzMng \ud83d\udd17Obscura Remova LoRA: https:\/\/huggingface.co\/WepeNerd\/Obscura_Remova \ud83d\udd17 Wan2GP repo: https:\/\/github.com\/deepbeepmeep\/Wan2GP \ud83d\udcdd Topics covered: LTX-2.3 LoRA workflow Obscura Remova demo Wan2GP tutorial AI video object removal Removing obstructions from video Foreground object removal with AI video LTX 2.3 is an open video generation model used for image-to-video, video-to-video and text-to-video. Wan2GP is an open-source app for running AI image and video models on consumer GPUs, including LTX Video. It&#039;s like an easier to use ComfyUI. #LTX23 #LTXVideo #ObjectRemoval #Wan2GP #AIVideo #VideoEditing #AITools #opensourceai #runpod #AIFilmmaking #AIVideoGeneration #LoRA\"><\/div><\/div><div class=\"lL\" style=\"max-width:100%;width:853px;margin:5px auto;\"><\/div><figcaption><\/figcaption><\/figure>\n\n<figure class=\"wp-block-embed-youtube wp-block-embed is-type-video is-provider-youtube wp-embed-aspect-16-9 wp-has-aspect-ratio\"><div class=\"lyte-wrapper\" title=\"\u3010ComfyUI\u3011\u5c0f\u5c3a\u5ea6\u5de5\u4f5c\u6d41\uff01\u628aLTX 2 3\u53c8\u69a8\u5e72\u4e86\u4e00\u6b21\uff1a\u89c6\u9891\u751f\u6210\u7ec8\u4e8e\u66f4\u7a33\u3001\u66f4\u51c6\u3001\u66f4\u542c\u8bdd\u4e86 p01 Ltx2 3\u7ee7\u7eed&ldquo;\u69a8\u5e72&rdquo;\" style=\"width:853px;max-width:100%;margin:5px auto;\"><div class=\"lyMe\" id=\"WYL_uQpRKwOjjZM\" itemprop=\"video\" itemscope itemtype=\"https:\/\/schema.org\/VideoObject\"><div><meta itemprop=\"thumbnailUrl\" content=\"https:\/\/infernews.com\/blog\/wp-content\/plugins\/wp-youtube-lyte\/lyteCache.php?origThumbUrl=https%3A%2F%2Fi.ytimg.com%2Fvi%2FuQpRKwOjjZM%2Fhqdefault.jpg\" \/><meta itemprop=\"embedURL\" content=\"https:\/\/www.youtube.com\/embed\/uQpRKwOjjZM\" \/><meta itemprop=\"duration\" content=\"PT4M19S\" \/><meta itemprop=\"uploadDate\" content=\"2026-05-22T02:36:43Z\" \/><\/div><div id=\"lyte_uQpRKwOjjZM\" data-src=\"https:\/\/infernews.com\/blog\/wp-content\/plugins\/wp-youtube-lyte\/lyteCache.php?origThumbUrl=https%3A%2F%2Fi.ytimg.com%2Fvi%2FuQpRKwOjjZM%2Fhqdefault.jpg\" class=\"pL\"><div class=\"tC\"><div class=\"tT\" itemprop=\"name\">\u3010ComfyUI\u3011\u5c0f\u5c3a\u5ea6\u5de5\u4f5c\u6d41\uff01\u628aLTX 2 3\u53c8\u69a8\u5e72\u4e86\u4e00\u6b21\uff1a\u89c6\u9891\u751f\u6210\u7ec8\u4e8e\u66f4\u7a33\u3001\u66f4\u51c6\u3001\u66f4\u542c\u8bdd\u4e86 p01 Ltx2 3\u7ee7\u7eed\u201c\u69a8\u5e72\u201d<\/div><\/div><button tabindex=\"0\" class=\"play\"><\/button><div class=\"ctrl\"><div class=\"Lctrl\"><\/div><div class=\"Rctrl\"><\/div><\/div><\/div><noscript><a href=\"https:\/\/youtu.be\/uQpRKwOjjZM\" rel=\"nofollow\"><img loading=\"lazy\" decoding=\"async\" src=\"https:\/\/infernews.com\/blog\/wp-content\/plugins\/wp-youtube-lyte\/lyteCache.php?origThumbUrl=https%3A%2F%2Fi.ytimg.com%2Fvi%2FuQpRKwOjjZM%2F0.jpg\" alt=\"\u3010ComfyUI\u3011\u5c0f\u5c3a\u5ea6\u5de5\u4f5c\u6d41\uff01\u628aLTX 2 3\u53c8\u69a8\u5e72\u4e86\u4e00\u6b21\uff1a\u89c6\u9891\u751f\u6210\u7ec8\u4e8e\u66f4\u7a33\u3001\u66f4\u51c6\u3001\u66f4\u542c\u8bdd\u4e86 p01 Ltx2 3\u7ee7\u7eed&ldquo;\u69a8\u5e72&rdquo;\" width=\"853\" height=\"460\" \/><br \/>Watch this video on YouTube<\/a><\/noscript><meta itemprop=\"description\" content=\"\u300cLTX\u63d2\u4ef6\u548c\u6a21\u578b\u5168\u5957\u300d\u94fe\u63a5\uff1ahttps:\/\/pan.quark.cn\/s\/58b98f6d688f \u300cComfyUI V8\u5b8c\u6574\u6574\u5408\u5305\u300d \u94fe\u63a5\uff1ahttps:\/\/pan.quark.cn\/s\/7d3a8c64d9c3 \u300c\u767e\u5b9d\u7bb1\u300d \u94fe\u63a5\uff1ahttps:\/\/pan.quark.cn\/s\/2bbad26833fc \u300cLTX2.3\u300d\u94fe\u63a5\uff1ahttps:\/\/pan.quark.cn\/s\/17beb2861b6b \u300cFlux2-Klein-Face-Swap\u6362\u8138\u300d\u94fe\u63a5\uff1ahttps:\/\/pan.quark.cn\/s\/54a58a5e0f62 \u300cFlux\u8d44\u6e90\u6574\u5408\u5305\u300d \u94fe\u63a5\uff1ahttps:\/\/pan.quark.cn\/s\/53adf7c46b99 SVIPro\u5355\u56fe\u63d0\u793a\u8bcd\u5168\u81ea\u52a8\u957f\u89c6\u9891\u751f\u6210 \u4e00\u952e\u542f\u52a8\u5305+\u6a21\u578b+\u5de5\u4f5c\u6d41\u94fe\u63a5\uff1ahttps:\/\/pan.quark.cn\/s\/e7e7b5e04a29 \u300cai\u4e16\u754cseedance2\u300d\u94fe\u63a5\uff1ahttps:\/\/pan.quark.cn\/s\/8b83afb0bc63 \u300cComfyUI-seedance2\u5de5\u4f5c\u6d41]:https:\/\/pan.quark.cn\/s\/118540347246 \u300cLTX2.0\u590d\u523bseedance2.0\u300d\u94fe\u63a5\uff1ahttps:\/\/pan.quark.cn\/s\/d58288786f22 \u841d\u535c\u5168\u90e8\u8d44\u6e90\uff1ahttps:\/\/pan.quark.cn\/s\/0a253b3f1d8c \u300csd-webui-aki\u300d\u6574\u5408\u5305\uff1ahttps:\/\/pan.quark.cn\/s\/345008fa69f0 WAN\u6574\u5408\u5305\u5e26\u6a21\u578b\u94fe\u63a5\uff1ahttps:\/\/pan.quark.cn\/s\/b355e6e86853 \u300cAI-Tools\u67e5\u627e\u65b0\u5de5\u5177\u6309\u65e5\u671f\uff0c\u67e5\u627e\u5bf9\u5e94\u5de5\u5177\u6309\u540d\u79f0\u6392\u5e8f\u300d \u94fe\u63a5\uff1ahttps:\/\/pan.quark.cn\/s\/efc430e1996d\"><\/div><\/div><div class=\"lL\" style=\"max-width:100%;width:853px;margin:5px auto;\"><\/div><figcaption><\/figcaption><\/figure>\n\n<figure class=\"wp-block-embed-youtube wp-block-embed is-type-video is-provider-youtube wp-embed-aspect-16-9 wp-has-aspect-ratio\"><div class=\"lyte-wrapper\" title=\"LTX 2.3 Fine-Tuned Models: Better Consistency, Motion &amp;amp; Prompt Control &mdash;Subtitle &amp;amp; Watermark Removal\" style=\"width:853px;max-width:100%;margin:5px auto;\"><div class=\"lyMe\" id=\"WYL_SQs9X2M9IVA\" itemprop=\"video\" itemscope itemtype=\"https:\/\/schema.org\/VideoObject\"><div><meta itemprop=\"thumbnailUrl\" content=\"https:\/\/infernews.com\/blog\/wp-content\/plugins\/wp-youtube-lyte\/lyteCache.php?origThumbUrl=https%3A%2F%2Fi.ytimg.com%2Fvi%2FSQs9X2M9IVA%2Fhqdefault.jpg\" \/><meta itemprop=\"embedURL\" content=\"https:\/\/www.youtube.com\/embed\/SQs9X2M9IVA\" \/><meta itemprop=\"duration\" content=\"PT24M18S\" \/><meta itemprop=\"uploadDate\" content=\"2026-05-21T13:30:18Z\" \/><\/div><div id=\"lyte_SQs9X2M9IVA\" data-src=\"https:\/\/infernews.com\/blog\/wp-content\/plugins\/wp-youtube-lyte\/lyteCache.php?origThumbUrl=https%3A%2F%2Fi.ytimg.com%2Fvi%2FSQs9X2M9IVA%2Fhqdefault.jpg\" class=\"pL\"><div class=\"tC\"><div class=\"tT\" itemprop=\"name\">LTX 2.3 Fine-Tuned Models: Better Consistency, Motion &amp; Prompt Control \u2014Subtitle &amp; Watermark Removal<\/div><\/div><button tabindex=\"0\" class=\"play\"><\/button><div class=\"ctrl\"><div class=\"Lctrl\"><\/div><div class=\"Rctrl\"><\/div><\/div><\/div><noscript><a href=\"https:\/\/youtu.be\/SQs9X2M9IVA\" rel=\"nofollow\"><img loading=\"lazy\" decoding=\"async\" src=\"https:\/\/infernews.com\/blog\/wp-content\/plugins\/wp-youtube-lyte\/lyteCache.php?origThumbUrl=https%3A%2F%2Fi.ytimg.com%2Fvi%2FSQs9X2M9IVA%2F0.jpg\" alt=\"LTX 2.3 Fine-Tuned Models: Better Consistency, Motion &amp;amp; Prompt Control &mdash;Subtitle &amp;amp; Watermark Removal\" width=\"853\" height=\"460\" \/><br \/>Watch this video on YouTube<\/a><\/noscript><meta itemprop=\"description\" content=\"After Wan stopped releasing open-source versions, LTX 2.3 became the latest and most powerful open-source video generation model available, and the open-source community has produced extensive fine-tunes around it. Back in the Wan 2.1 era, we had FusionX, which delivered better performance and results. Right now, of course, we also have fine-tuned versions of LTX 2.3. In today&#039;s video, I&#039;ll walk through several distinct and worth-trying fine-tuned models based on LTX 2.3 \u2014 Sulphur-2, 10Eros, and ICEdit-Insight. Models\uff1a Sulphur - 2: https:\/\/huggingface.co\/SulphurAI\/Sulphur-2-base\/tree\/main Sulphur - 2 GGUF Version: https:\/\/huggingface.co\/Abiray\/Sulphur-2-base-GGUF\/tree\/main 10Eros: https:\/\/huggingface.co\/TenStrip\/LTX2.3-10Eros\/tree\/main 10Eros-GGUF Version\uff1ahttps:\/\/huggingface.co\/Abiray\/LTX2.3-10Eros-GGUF\/tree\/main ICEdit-Insight https:\/\/huggingface.co\/joyfox\/LTX2.3-ICEdit-Insight\/tree\/main CustomNodes\uff1a 10S-Comfy-nodes: https:\/\/github.com\/TenStrip\/10S-Comfy-nodes KJ Nodes\uff1ahttps:\/\/github.com\/kijai\/ComfyUI-KJNodes Mie Nodes\uff1ahttps:\/\/github.com\/MieMieeeee\/ComfyUI-MieNodes Impact Pack\uff1ahttps:\/\/github.com\/ltdrdata\/ComfyUI-Impact-Pack RES4LYF\uff1ahttps:\/\/github.com\/ClownsharkBatwing\/RES4LYF ComfyUI-LTXVideo\uff1ahttps:\/\/github.com\/Lightricks\/ComfyUI-LTXVide Workflow: https:\/\/drive.google.com\/drive\/folders\/1bdtYD6Ry-wDr_8B_JRrKAAdTSREWygT-?usp=sharing 00:00 \u2014 Effect showcase 01:33 \u2014 Sulphur-2 model intro and walkthrough 14:11 \u2014 10Eros model intro and walkthrough 19:42 \u2014 ICEdit-Insight model intro and walkthrough\"><\/div><\/div><div class=\"lL\" style=\"max-width:100%;width:853px;margin:5px auto;\"><\/div><figcaption><\/figcaption><\/figure>\n\n<figure class=\"wp-block-embed-youtube wp-block-embed is-type-video is-provider-youtube wp-embed-aspect-16-9 wp-has-aspect-ratio\"><div class=\"lyte-wrapper\" title=\"LTX2.3 Music Video Creator V5.1- ComfyUI - Full Walkthrough\" style=\"width:853px;max-width:100%;margin:5px auto;\"><div class=\"lyMe\" id=\"WYL_LwG-zxY684M\" itemprop=\"video\" itemscope itemtype=\"https:\/\/schema.org\/VideoObject\"><div><meta itemprop=\"thumbnailUrl\" content=\"https:\/\/infernews.com\/blog\/wp-content\/plugins\/wp-youtube-lyte\/lyteCache.php?origThumbUrl=https%3A%2F%2Fi.ytimg.com%2Fvi%2FLwG-zxY684M%2Fhqdefault.jpg\" \/><meta itemprop=\"embedURL\" content=\"https:\/\/www.youtube.com\/embed\/LwG-zxY684M\" \/><meta itemprop=\"duration\" content=\"PT18M26S\" \/><meta itemprop=\"uploadDate\" content=\"2026-05-08T23:43:16Z\" \/><\/div><div id=\"lyte_LwG-zxY684M\" data-src=\"https:\/\/infernews.com\/blog\/wp-content\/plugins\/wp-youtube-lyte\/lyteCache.php?origThumbUrl=https%3A%2F%2Fi.ytimg.com%2Fvi%2FLwG-zxY684M%2Fhqdefault.jpg\" class=\"pL\"><div class=\"tC\"><div class=\"tT\" itemprop=\"name\">LTX2.3 Music Video Creator V5.1- ComfyUI - Full Walkthrough<\/div><\/div><button tabindex=\"0\" class=\"play\"><\/button><div class=\"ctrl\"><div class=\"Lctrl\"><\/div><div class=\"Rctrl\"><\/div><\/div><\/div><noscript><a href=\"https:\/\/youtu.be\/LwG-zxY684M\" rel=\"nofollow\"><img loading=\"lazy\" decoding=\"async\" src=\"https:\/\/infernews.com\/blog\/wp-content\/plugins\/wp-youtube-lyte\/lyteCache.php?origThumbUrl=https%3A%2F%2Fi.ytimg.com%2Fvi%2FLwG-zxY684M%2F0.jpg\" alt=\"LTX2.3 Music Video Creator V5.1- ComfyUI - Full Walkthrough\" width=\"853\" height=\"460\" \/><br \/>Watch this video on YouTube<\/a><\/noscript><meta itemprop=\"description\" content=\"Create automated AI music videos with my full LTX 2.3 workflow for ComfyUI FREE and LOCAL! In this walkthrough, I show how the workflow takes a song, analyzes the timing, creates scene prompts from lyrics, and generates a finished music video using LTX 2.3. The workflow is split into two parts. \ud83c\udfb5 Workflow 1 handles audio upload, beat detection, scene timing, lyrics, style and theme, story idea, subjects and locations, and prompt generation. \ud83c\udfac Workflow 2 handles the actual video generation, including an image-to-video with zImage Turbo and LTX 2.3 workflow, and a text-to-video with LTX, LoRa workflow that both have advanced prompt controls, scene generation, Remake Mode, and final video stitching. \u2728 This workflow is designed to reduce manual setup time while still giving you control over style, characters, camera motion, timing, seeds, LoRas, and final edits. \ud83d\udca1 For the best results, I recommend starting with the default settings first, then experimenting with LoRas, seeds, advanced settings, and Remake Mode as you get more comfortable. \u2699\ufe0f Requirements: ComfyUI LTX 2.3 models Z-Image Turbo model FFmpeg installed for audio stitching My vrgamedevgirl custom nodes Impact Pack custom node for auto-queue llama-cpp-python 16 GB of VRAM min recommended \ud83d\udcac Join my Discord for support, updates, beta features, and to share your work: https:\/\/discord.gg\/ltx23musicvideocreator \u2b07\ufe0f Download my custom nodes and workflows: https:\/\/github.com\/vrgamegirl19\/comfyui-vrgamedevgirl #ComfyUI #LTX #AIvideo #AIMusicVideo #TextToVideo #ImageToVideo #AIWorkflow #GenerativeAI\"><\/div><\/div><div class=\"lL\" style=\"max-width:100%;width:853px;margin:5px auto;\"><\/div><figcaption><\/figcaption><\/figure>\n\n\n<h2 class=\"wp-block-heading\">Reference Sheet with KREA<\/h2>\n\n\n<figure class=\"wp-block-embed-youtube wp-block-embed is-type-video is-provider-youtube wp-embed-aspect-16-9 wp-has-aspect-ratio\"><div class=\"lyte-wrapper\" title=\"This Workflow Makes AI Influencers Consistent &mdash; No LoRA Training\" style=\"width:853px;max-width:100%;margin:5px auto;\"><div class=\"lyMe\" id=\"WYL_OCvTf34RlAk\" itemprop=\"video\" itemscope itemtype=\"https:\/\/schema.org\/VideoObject\"><div><meta itemprop=\"thumbnailUrl\" content=\"https:\/\/infernews.com\/blog\/wp-content\/plugins\/wp-youtube-lyte\/lyteCache.php?origThumbUrl=https%3A%2F%2Fi.ytimg.com%2Fvi%2FOCvTf34RlAk%2Fhqdefault.jpg\" \/><meta itemprop=\"embedURL\" content=\"https:\/\/www.youtube.com\/embed\/OCvTf34RlAk\" \/><meta itemprop=\"duration\" content=\"PT10M16S\" \/><meta itemprop=\"uploadDate\" content=\"2026-08-02T13:16:56Z\" \/><\/div><div id=\"lyte_OCvTf34RlAk\" data-src=\"https:\/\/infernews.com\/blog\/wp-content\/plugins\/wp-youtube-lyte\/lyteCache.php?origThumbUrl=https%3A%2F%2Fi.ytimg.com%2Fvi%2FOCvTf34RlAk%2Fhqdefault.jpg\" class=\"pL\"><div class=\"tC\"><div class=\"tT\" itemprop=\"name\">This Workflow Makes AI Influencers Consistent \u2014 No LoRA Training<\/div><\/div><button tabindex=\"0\" class=\"play\"><\/button><div class=\"ctrl\"><div class=\"Lctrl\"><\/div><div class=\"Rctrl\"><\/div><\/div><\/div><noscript><a href=\"https:\/\/youtu.be\/OCvTf34RlAk\" rel=\"nofollow\"><img loading=\"lazy\" decoding=\"async\" src=\"https:\/\/infernews.com\/blog\/wp-content\/plugins\/wp-youtube-lyte\/lyteCache.php?origThumbUrl=https%3A%2F%2Fi.ytimg.com%2Fvi%2FOCvTf34RlAk%2F0.jpg\" alt=\"This Workflow Makes AI Influencers Consistent &mdash; No LoRA Training\" width=\"853\" height=\"460\" \/><br \/>Watch this video on YouTube<\/a><\/noscript><meta itemprop=\"description\" content=\"Stop training LoRAs every time you want a new AI influencer. In this video I show my ComfyUI Reference Influencer workflow \u2014 scene photo + character sheet in, polished still out \u2014 using Krea 2 Edit + Identity Edit LoRA. Free &amp; local. Optional SeedVR sharpen. Same graph can flip into LTX 2.3 video later. You&#039;ll learn: \u2022 Dual-image edit: IMAGE 1 = scene, IMAGE 2 = character sheet \/ portrait \u2022 Krea 2 Edit + Identity Edit v1.2 (no dedicated character LoRA training) \u2022 PureLens \/ JPEG Core \/ Influencer LoRA stack for that camera-real look \u2022 Controls that matter (strengths, SeedVR, when to skip upscale) \u2022 How to keep face + outfit locked while changing the background \u2b07 DOWNLOAD THIS WORKFLOW (pay what you want) https:\/\/shahzaib632.gumroad.com\/l\/nzdpsq Includes: \u2022 Workflow JSON \u2014 Reference Influencer + Video (Krea2 Edit + LTX23) \u2022 SETUP_GUIDE.md \u2014 exact models, LoRAs, nodes, folders, versions \u2022 5 sample character sheets + sheet-creation prompt Related workflows: \u2022 Krea 2 + LTX 2.3 video realism (Part 2): https:\/\/www.youtube.com\/watch?v=rI0n7ly-0xA \u2022 Krea 2 stills that look scary real (Part 1): https:\/\/www.youtube.com\/watch?v=38GPmAVL3lc \u2022 Free Krea 2 Realism workflow: https:\/\/shahzaib632.gumroad.com\/l\/lnvyhi \u2022 Celebrity voice + lipsync: https:\/\/shahzaib632.gumroad.com\/l\/ukbmw \u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500 HOW THE WORKFLOW WORKS \u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500 1. Install custom nodes from the SETUP_GUIDE (or list below) 2. Download models into the exact folders listed 3. Open the JSON in ComfyUI \u2014 zero red nodes 4. Load SCENE into image 1, CHARACTER into image 2 (crop sheet to close-up) 5. Write an Edit Instruction (start with \u201cpurelens.\u201d) 6. Keep Image to Video OFF \u2192 Queue \u2192 judge the still 7. Optional SeedVR on the still (skip if skin looks too smooth) 8. When ready for motion, flip Image to Video ON (covered in Part 2) \u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500 MODELS USED (download these) \u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500 KREA 2 EDIT (still \/ identity) \u2022 Krea 2 Turbo fp8 UNET: https:\/\/huggingface.co\/Comfy-Org\/Krea-2\/resolve\/main\/diffusion_models\/krea2_turbo_fp8_scaled.safetensors \u2022 Qwen3-VL 4B Text Encoder (CLIP type = krea2): https:\/\/huggingface.co\/Comfy-Org\/Krea-2\/resolve\/main\/text_encoders\/qwen3vl_4b_fp8_scaled.safetensors \u2022 Qwen Image VAE: https:\/\/huggingface.co\/Comfy-Org\/Krea-2\/resolve\/main\/vae\/qwen_image_vae.safetensors KREA LORAS \u2022 Identity Edit v1.2 (REQUIRED @ 1.0) \u2192 put in loras\/Krea2\/: https:\/\/huggingface.co\/conradlocke\/krea2-identity-edit\/resolve\/main\/krea2_identity_edit_v1_2.safetensors \u2022 PureLens Realism Krea2 @ 0.65 (trigger: purelens): https:\/\/civitai.com\/model-versions\/3114242 \u2022 JPEG Core Krea2 @ 0.40 (lower to 0.1\u20130.2 if soft): https:\/\/civitai.com\/model-versions\/3113304 \u2022 Instagram Influencers Krea2 @ 0.45: https:\/\/civitai.com\/model-versions\/3081379 \u2022 Realism Slider v1 @ 0.30: https:\/\/civitai.com\/model-versions\/3132956 SEEDVR2 (optional still upscale) \u2022 SeedVR2 DiT 3B fp8: https:\/\/huggingface.co\/numz\/SeedVR2_comfyUI\/resolve\/main\/seedvr2_ema_3b_fp8_e4m3fn.safetensors \u2022 SeedVR2 VAE fp16: https:\/\/huggingface.co\/numz\/SeedVR2_comfyUI\/resolve\/main\/ema_vae_fp16.safetensors LTX 2.3 (optional video branch \u2014 same pack) \u2022 LTX 2.3 22B Dev FP8: https:\/\/huggingface.co\/Lightricks\/LTX-2.3-fp8\/resolve\/main\/ltx-2.3-22b-dev-fp8.safetensors \u2022 Gemma 3 12B IT TE fp8: https:\/\/huggingface.co\/Comfy-Org\/ltx-2\/resolve\/main\/split_files\/text_encoders\/gemma_3_12B_it_fp8_scaled.safetensors \u2022 Distilled LoRA 384 v1.1 \u2192 must be in loras\/ltx2\/: https:\/\/huggingface.co\/Lightricks\/LTX-2.3\/resolve\/main\/ltx-2.3-22b-distilled-lora-384-1.1.safetensors \u2022 Spatial upscaler x2 v1.1: https:\/\/huggingface.co\/Lightricks\/LTX-2.3\/resolve\/main\/ltx-2.3-spatial-upscaler-x2-1.1.safetensors \u2022 Full LTX 2.3 model card: https:\/\/huggingface.co\/Lightricks\/LTX-2.3 \u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500 CUSTOM NODES REQUIRED \u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500 \u2022 comfyui-krea2edit (v1.2+): https:\/\/github.com\/lbouaraba\/comfyui-krea2edit \u2022 ComfyUI-LTXVideo: https:\/\/github.com\/Lightricks\/ComfyUI-LTXVideo \u2022 ComfyUI-KJNodes: https:\/\/github.com\/kijai\/ComfyUI-KJNodes \u2022 10S \/ TenStrip: https:\/\/github.com\/TenStrip\/10S-Comfy-nodes \u2022 SeedVR2: https:\/\/github.com\/numz\/ComfyUI-SeedVR2_VideoUpscaler \u2022 VideoHelperSuite: https:\/\/github.com\/Kosinkadink\/ComfyUI-VideoHelperSuite \u2022 rgthree-comfy: https:\/\/github.com\/rgthree\/rgthree-comfy \u2022 Easy-Use \u00b7 mxToolkit \u00b7 ComfyMath (Manager search) TIMESTAMPS: 0:00 Result first 0:29 Workflow Walkthrough 4:08 Demos 9:23 Download Page &amp; Walkthrough If this helped, like + subscribe for more free local ComfyUI workflows. Comment: still training character LoRAs, or switching to reference sheets? https:\/\/www.youtube.com\/@TheAIBriefYT20 #ComfyUI #Krea2 #AIInfluencer #ConsistentCharacter #LocalAI #SeedVR2 #AIworkflow\"><\/div><\/div><div class=\"lL\" style=\"max-width:100%;width:853px;margin:5px auto;\"><\/div><figcaption><\/figcaption><\/figure>\n\n\n<hr class=\"wp-block-separator has-alpha-channel-opacity\"\/>\n\n\n\n<h2 class=\"wp-block-heading\">\u5b78\u7fd2<\/h2>\n\n\n<figure class=\"wp-block-embed-youtube wp-block-embed is-type-video is-provider-youtube wp-embed-aspect-16-9 wp-has-aspect-ratio\"><div class=\"lyte-wrapper\" title=\"\u7206\u706b\u7684&ldquo;\u65e0\u5ba1\u67e5&rdquo;AI \u89c6\u9891\u6a21\u578b\u6765\u4e86\uff01Sulphur 2 \u672c\u5730\u90e8\u7f72\u5b9e\u6d4b\uff1a8G \u663e\u5b58\u4e5f\u80fd\u8dd1\uff01\u5b8c\u5168\u514d\u8d39\u5f00\u6e90 | \u96f6\u5ea6\u89e3\u8bf4\" style=\"width:853px;max-width:100%;margin:5px auto;\"><div class=\"lyMe\" id=\"WYL_Nms56KhvIQw\" itemprop=\"video\" itemscope itemtype=\"https:\/\/schema.org\/VideoObject\"><div><meta itemprop=\"thumbnailUrl\" content=\"https:\/\/infernews.com\/blog\/wp-content\/plugins\/wp-youtube-lyte\/lyteCache.php?origThumbUrl=https%3A%2F%2Fi.ytimg.com%2Fvi%2FNms56KhvIQw%2Fhqdefault.jpg\" \/><meta itemprop=\"embedURL\" content=\"https:\/\/www.youtube.com\/embed\/Nms56KhvIQw\" \/><meta itemprop=\"duration\" content=\"PT10M11S\" \/><meta itemprop=\"uploadDate\" content=\"2026-05-11T13:12:31Z\" \/><\/div><div id=\"lyte_Nms56KhvIQw\" data-src=\"https:\/\/infernews.com\/blog\/wp-content\/plugins\/wp-youtube-lyte\/lyteCache.php?origThumbUrl=https%3A%2F%2Fi.ytimg.com%2Fvi%2FNms56KhvIQw%2Fhqdefault.jpg\" class=\"pL\"><div class=\"tC\"><div class=\"tT\" itemprop=\"name\">\u7206\u706b\u7684\u201c\u65e0\u5ba1\u67e5\u201dAI \u89c6\u9891\u6a21\u578b\u6765\u4e86\uff01Sulphur 2 \u672c\u5730\u90e8\u7f72\u5b9e\u6d4b\uff1a8G \u663e\u5b58\u4e5f\u80fd\u8dd1\uff01\u5b8c\u5168\u514d\u8d39\u5f00\u6e90 | \u96f6\u5ea6\u89e3\u8bf4<\/div><\/div><button tabindex=\"0\" class=\"play\"><\/button><div class=\"ctrl\"><div class=\"Lctrl\"><\/div><div class=\"Rctrl\"><\/div><\/div><\/div><noscript><a href=\"https:\/\/youtu.be\/Nms56KhvIQw\" rel=\"nofollow\"><img loading=\"lazy\" decoding=\"async\" src=\"https:\/\/infernews.com\/blog\/wp-content\/plugins\/wp-youtube-lyte\/lyteCache.php?origThumbUrl=https%3A%2F%2Fi.ytimg.com%2Fvi%2FNms56KhvIQw%2F0.jpg\" alt=\"\u7206\u706b\u7684&ldquo;\u65e0\u5ba1\u67e5&rdquo;AI \u89c6\u9891\u6a21\u578b\u6765\u4e86\uff01Sulphur 2 \u672c\u5730\u90e8\u7f72\u5b9e\u6d4b\uff1a8G \u663e\u5b58\u4e5f\u80fd\u8dd1\uff01\u5b8c\u5168\u514d\u8d39\u5f00\u6e90 | \u96f6\u5ea6\u89e3\u8bf4\" width=\"853\" height=\"460\" \/><br \/>Watch this video on YouTube<\/a><\/noscript><meta itemprop=\"description\" content=\"\u3010\u66f4\u591a\u8d44\u6e90\u3011\u25b6https:\/\/bittly.cc\/lingdu \u3010\u96f6\u5ea6\u535a\u5ba2\u3011\u25b6https:\/\/www.freedidi.com \u3010\u52a0\u5165\u4f1a\u5458\u3011\u25b6https:\/\/www.youtube.com\/channel\/UCvijahEyGtvMpmMHBu4FS2w\/join \u3010\u9ad8\u7ea7\u4f1a\u5458\u3011\u25b6https:\/\/bittly.cc\/JsLAn ------------------------------------- \ud83d\udd14 Sulphur 2 \u5f00\u6e90\u6a21\u578b\u4e0b\u8f7d\uff1ahttps:\/\/www.freedidi.com\/24142.html \ud83e\udeb6\ud83e\udd47\ud83e\udd48\ud83e\udd49 \u4f1a\u5458\u54a8\u8be2 \/ \u4e00\u5bf9\u4e00\uff08\u9650\u91cf\u5f00\u653e\uff09\uff1a\ud83d\udd17https:\/\/ask.tuio.cc \ud83d\udd30 \u5b89\u5168VPN\u63a8\u8350\uff08\u6700\u9ad8\u4f18\u60e0\u94fe\u63a5\uff09\uff1a 1.SurfsharkVPN\uff1ahttps:\/\/get.surfshark.net\/aff_c?offer_id=390&amp;aff_id=4890 2.ProtonVPN\uff1ahttps:\/\/go.getproton.me\/aff_c?offer_id=26&amp;aff_id=1905&amp;source=youtube 3.ExpressVPN\uff1ahttps:\/\/go.expressvpn.com\/c\/2612933\/1645830\/16063 4.PureVPN\uff1a https:\/\/billing.purevpn.com\/aff.php?aff=44074 \ud83d\ude80VPS \u63a8\u8350\uff1a 1.https:\/\/bandwagonhost.com\/aff.php?aff=58794 \uff08\u642c\u74e6\u5de5 CN2\uff09 \ud83e\udd9e AI \u751f\u6210\u5bf9\u53e3\u578b\u89c6\u9891\uff08\u652f\u6301\u591a\u8bed\u8a00\uff09\uff1ahttps:\/\/bittly.cc\/heygen \ud83d\udcb0\u6bd4\u7279\u5e01\u4ea4\u6613\u5e73\u53f0 1.\u706b\u5e01\u5b98\u7f51 : https:\/\/www.htx.com\/invite\/zh-cn\/1f?invite_code=48tr7 2.\u6b27\u6613\u5b98\u7f51: https:\/\/www.okx.com\/join\/15073927 ======================= \u5176\u5b83\u66f4\u591a\u7cbe\u5f69\u5185\u5bb9\u63a8\u8350\uff1a\ud83d\ude0d https:\/\/youtu.be\/-VrgBEz-EZ4 https:\/\/youtu.be\/bowzVW93pig https:\/\/youtu.be\/uFHtS_kJAUE https:\/\/youtu.be\/dNsxnuAIIdw https:\/\/youtu.be\/q1qCzx3sakA https:\/\/youtu.be\/swafOCGzmow https:\/\/youtu.be\/5eA7-gWH_2o https:\/\/youtu.be\/HAXed-ZYEZQ https:\/\/youtu.be\/nc-EQHtx2W8 https:\/\/youtu.be\/LQwD0x1Z8ak https:\/\/youtu.be\/T4eLUtIRy_M https:\/\/youtu.be\/5XrayNHLWrg https:\/\/youtu.be\/b_NTQEWJLxw https:\/\/youtu.be\/lzb5bJS4XtI https:\/\/youtu.be\/cHIyeTdKB4g https:\/\/youtu.be\/p8KTaBv7nxw https:\/\/youtu.be\/UWelKj2LkIU https:\/\/youtu.be\/JBv-thjmVP8 https:\/\/youtu.be\/QCvF8mCPwdQ https:\/\/youtu.be\/NLO0nx4EN24 https:\/\/youtu.be\/r351ngmDL3Q https:\/\/youtu.be\/IIcAe2AlROE https:\/\/youtu.be\/kCrX3Dm4_cY https:\/\/youtu.be\/UhMim4G79Oo https:\/\/youtu.be\/xzBiF8TJ_t4 https:\/\/youtu.be\/Q-FUmOtysyg https:\/\/youtu.be\/MEZnK8ZOSVw ========================= \u96f6\u5ea6\u7535\u62a5\u7fa4\uff1ahttps:\/\/t.me\/lingdujie\"><\/div><\/div><div class=\"lL\" style=\"max-width:100%;width:853px;margin:5px auto;\"><\/div><figcaption><\/figcaption><\/figure>\n\n<figure class=\"wp-block-embed-youtube wp-block-embed is-type-video is-provider-youtube wp-embed-aspect-16-9 wp-has-aspect-ratio\"><div class=\"lyte-wrapper\" title=\"Advanced ComfyUI Nodes That Instantly Upgrade Your Workflow\" style=\"width:853px;max-width:100%;margin:5px auto;\"><div class=\"lyMe\" id=\"WYL_yfN-DMCoue0\" itemprop=\"video\" itemscope itemtype=\"https:\/\/schema.org\/VideoObject\"><div><meta itemprop=\"thumbnailUrl\" content=\"https:\/\/infernews.com\/blog\/wp-content\/plugins\/wp-youtube-lyte\/lyteCache.php?origThumbUrl=https%3A%2F%2Fi.ytimg.com%2Fvi%2FyfN-DMCoue0%2Fhqdefault.jpg\" \/><meta itemprop=\"embedURL\" content=\"https:\/\/www.youtube.com\/embed\/yfN-DMCoue0\" \/><meta itemprop=\"duration\" content=\"PT27M17S\" \/><meta itemprop=\"uploadDate\" content=\"2026-07-14T21:34:26Z\" \/><\/div><div id=\"lyte_yfN-DMCoue0\" data-src=\"https:\/\/infernews.com\/blog\/wp-content\/plugins\/wp-youtube-lyte\/lyteCache.php?origThumbUrl=https%3A%2F%2Fi.ytimg.com%2Fvi%2FyfN-DMCoue0%2Fhqdefault.jpg\" class=\"pL\"><div class=\"tC\"><div class=\"tT\" itemprop=\"name\">Advanced ComfyUI Nodes That Instantly Upgrade Your Workflow<\/div><\/div><button tabindex=\"0\" class=\"play\"><\/button><div class=\"ctrl\"><div class=\"Lctrl\"><\/div><div class=\"Rctrl\"><\/div><\/div><\/div><noscript><a href=\"https:\/\/youtu.be\/yfN-DMCoue0\" rel=\"nofollow\"><img loading=\"lazy\" decoding=\"async\" src=\"https:\/\/infernews.com\/blog\/wp-content\/plugins\/wp-youtube-lyte\/lyteCache.php?origThumbUrl=https%3A%2F%2Fi.ytimg.com%2Fvi%2FyfN-DMCoue0%2F0.jpg\" alt=\"Advanced ComfyUI Nodes That Instantly Upgrade Your Workflow\" width=\"853\" height=\"460\" \/><br \/>Watch this video on YouTube<\/a><\/noscript><meta itemprop=\"description\" content=\"\ud83c\udfac Advanced ComfyUI Nodes That Instantly Upgrade Your Workflow These are my top 7 favorite ComfyUI custom nodes as of July 2026. Not a most downloaded list, not sponsored \u2014 just the nodes I actually run in every single workflow right now. I built one workflow using all 7 and showcase each one live so you can see exactly what they do and how to dial them in. Node packs used: Anything Everywhere \u2192 install via ComfyUI Manager: cg-use-everywhere CR Prompt List \u2192 install via ComfyUI Manager: ComfyUI_Comfyroll_CustomNodes IdentityForge \u2192 install via ComfyUI Manager: comfyui-identity-forge SeedVarianceEnhancer \u2192 install via ComfyUI Manager: SeedVarianceEnhancer TextGenerate \/ Gemma 4 \u2192 built into ComfyUI Core Expression Editor \u2192 install via ComfyUI Manager: ComfyUI-AdvancedLivePortrait FaceDetailer \u2192 install via ComfyUI Manager: ComfyUI-Impact-Pack \ud83d\udd17 LINKS Discord (free workflows + 1 on 1 coaching): https:\/\/discord.gg\/2bqQXKxsxg \u2014 coaching covers ComfyUI\/RunPod setup, LoRA training, and AI influencer growth strategy. Consultation-based pricing. Details in the server. RunPod (GPU rental): https:\/\/runpod.io?ref=jc6190fx \u2014 add a minimum of &#36;10 through the link and get &#36;5 free. \ud83c\udff7\ufe0f TAGS comfyui, comfyui custom nodes, comfyui nodes 2026, best comfyui nodes, comfyui tutorial, comfyui workflow, comfyui tips, comfyui facedetailer, comfyui expression editor, comfyui gemma4, comfyui identity forge, comfyui seed variance, comfyui prompt list, comfyui anything everywhere, ai image generation, ai workflow, comfyui advanced, comfyui must have nodes, comfyui node pack, flux comfyui, ai content creation, comfyui 2026 #ComfyUI #AIImageGeneration #ComfyUINodes #AIWorkflow #ComfyUITutorial #Flux #AIContentCreation #ComfyUITips #GenerativeAI #AIArt #ComfyUICustomNodes #StableDiffusion #AITools #ComfyUI2026 #FaceDetailer\"><\/div><\/div><div class=\"lL\" style=\"max-width:100%;width:853px;margin:5px auto;\"><\/div><figcaption><\/figcaption><\/figure>\n\n\n<hr class=\"wp-block-separator has-alpha-channel-opacity\"\/>\n","protected":false},"excerpt":{"rendered":"<p>LTX 2.5 \u5feb\u901f\u4e0a\u624b All in one MSR LTX Trainer LTX Director Cross View Camera Control LTX IC-LoRA LTX 2.3 Reference Sheet with KREA \u5b78\u7fd2<\/p>\n","protected":false},"author":1,"featured_media":0,"parent":0,"menu_order":0,"comment_status":"closed","ping_status":"closed","template":"","meta":{"wpai_generated_summary":"","footnotes":""},"class_list":["post-8127","page","type-page","status-publish","hentry"],"_links":{"self":[{"href":"https:\/\/infernews.com\/blog\/wp-json\/wp\/v2\/pages\/8127","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/infernews.com\/blog\/wp-json\/wp\/v2\/pages"}],"about":[{"href":"https:\/\/infernews.com\/blog\/wp-json\/wp\/v2\/types\/page"}],"author":[{"embeddable":true,"href":"https:\/\/infernews.com\/blog\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/infernews.com\/blog\/wp-json\/wp\/v2\/comments?post=8127"}],"version-history":[{"count":1,"href":"https:\/\/infernews.com\/blog\/wp-json\/wp\/v2\/pages\/8127\/revisions"}],"predecessor-version":[{"id":11263,"href":"https:\/\/infernews.com\/blog\/wp-json\/wp\/v2\/pages\/8127\/revisions\/11263"}],"wp:attachment":[{"href":"https:\/\/infernews.com\/blog\/wp-json\/wp\/v2\/media?parent=8127"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}