<p>A simple, color-coded GGUF workflow for running LTXV-2 in ComfyUI.<br />Designed to be straightforward and lightweight.</p><div data-youtube-video><iframe width="640" height="480" allowfullscreen="true" autoplay="false" disablekbcontrols="false" enableiframeapi="false" endtime="0" ivloadpolicy="0" loop="false" modestbranding="false" origin playlist src="https://www.youtube.com/embed/im4wolfHvMk" start="0"></iframe></div><h3 id="common-issues"><strong>Common Issues</strong></h3><p><strong>Error:</strong> <code>LTXVEmptyLatentAudio – AttributeError</code></p><p>Update Kijai Node Pack to v1.2.8</p><p><strong>Checkerboard / patterned noise output</strong></p><p>Update all nodes + ComfyUI<br /><em>(Outdated GGUF loader or ComfyUI build)</em></p><hr /><h3 id="required-node-packs"><strong>Required Node Packs</strong></h3><p>ComfyUI-GGUF<br /><a target="_blank" rel="ugc" href="https://github.com/city96/ComfyUI-GGUF">https://github.com/city96/ComfyUI-GGUF</a></p><p>ComfyUI-KJNodes<br /><a target="_blank" rel="ugc" href="https://github.com/kijai/ComfyUI-KJNodes">https://github.com/kijai/ComfyUI-KJNodes</a> <em>(install &amp; update it to nightly)</em></p><p><em>additional install via manager</em></p><hr /><h1 id="downloads">📦 Downloads</h1><h2 id="kijai's-repo-(all-files-except-diffusion-models-choose-one-lora)">🧩 Kijai’s Repo (ALL files except diffusion models — choose ONE LoRA)</h2><p><a target="_blank" rel="ugc" href="https://huggingface.co/Kijai/LTX2.3_comfy/tree/main">https://huggingface.co/Kijai/LTX2.3_comfy/tree/main</a></p><p><strong>Includes:</strong></p><ul><li><p>Distilled LoRA (<strong>REQUIRED</strong>)</p></li><li><p>Text Projection (embedding connector)</p></li></ul><hr /><h2 id="diffusion-models-(gguf)">🧠 Diffusion Models (GGUF)</h2><p><a target="_blank" rel="ugc" href="https://huggingface.co/unsloth/LTX-2.3-GGUF/tree/main">https://huggingface.co/unsloth/LTX-2.3-GGUF/tree/main</a></p><p><strong>Choose based on VRAM:</strong></p><ul><li><p><code>Q4_K_M</code> → lower VRAM (<strong>recommended</strong>)</p></li><li><p><code>Q8_0</code> → higher quality, more VRAM</p></li></ul><p>⚠️ Very low quants = basically unusable</p><p><strong>Includes:</strong></p><ul><li><p>VAE (audio + video)</p></li><li><p>GGUF models</p></li></ul><hr /><h2 id="text-encoders-(required)">✍️ Text Encoders (<strong>REQUIRED</strong>)</h2><h3 id="option-1-gemma-(recommended-for-low-vram)">Option 1 – Gemma (recommended for low VRAM)</h3><p><a target="_blank" rel="ugc" href="https://huggingface.co/Comfy-Org/ltx-2/tree/main/split_files/text_encoders">https://huggingface.co/Comfy-Org/ltx-2/tree/main/split_files/text_encoders</a></p><h3 id="option-2-full-version-(higher-vram-better-quality)">Option 2 – Full version (higher VRAM, better quality)</h3><p>➡️ You still ALSO need:</p><ul><li><p>Text Projection (from Kijai’s repo)</p></li></ul><hr /><h2 id="latent-upscaler">🔍 Latent Upscaler</h2><p><a target="_blank" rel="ugc" href="https://huggingface.co/Lightricks/LTX-2/tree/main/">https://huggingface.co/Lightricks/LTX-2/tree/main/</a></p><p><strong>Files:</strong></p><ul><li><p><code>ltx-2-spatial-upscaler-x2-1.0.safetensors</code></p></li><li><p><code>ltx-2-temporal-upscaler-x2-1.0.safetensors</code></p></li></ul><hr /><h1 id="important-notes">⚠️ Important Notes</h1><ul><li><p>Use <strong>DEV GGUF diffusion models</strong> (no distilled base model yet)</p></li><li><p><strong>Distilled LoRA is still REQUIRED</strong></p></li><li><p>Most issues = outdated nodes or ComfyUI</p></li></ul><p>.</p><hr /><p></p>

LTXV2.3 GGUF I2V/T2V/FFLF+IDaudioLora 12gb Vram sageattn

ltxv23GGUFI2VT2VFFLF_v30T2VI2VFFLFIdlora.zip

Files with SHA256: e1d9aee9fd1055f0e5efd12e3429e54b06e643ddc1d27d56a4d099ea956aea89