<h1 id="ltx-see-motion-image-to-video-workflow-(ltx-2.3)">🎥 LTX See Motion — Image-to-Video Workflow (LTX-2.3)</h1>Two-stage ComfyUI workflow for the custom LTX See Motion LTX-2.3 checkpoint. Turns a single input image into a video with natural motion, synchronized audio and dialogue — identity, style and composition of the source image survive the whole clip.How it works: 🎬 Stage 1 renders the base clip (8 steps, CFG 1.0) → 🆙 Stage 2 upscales the latent 2x and refines it (4 steps, denoise 0.42) → ⚡ RTX Video SR sharpens another 1.5x → 💾 the final MP4 (H.264 + audio) saves automatically. Live low-VRAM preview included. Everything is grouped, color-coded and documented with README notes inside the workflow.<ul><li>Ships in FAST mode: distilled LoRA at strength 0.8, 8 + 4 steps, CFG 1.0</li><li>DEV mode: disable the LoRA, set Stage 1 to 40 steps / CFG 4.0 (inside the 🎬/🆙 stage subgraphs)</li><li>Strong: image-to-video (SFW + NSFW), dancing/singing, camera moves. Weak: plain text-to-video — always feed an image</li><li>Performance reference: RTX 5060 Ti 16 GB ≈ 300–400 s for a 15 s clip at ~1 MP</li></ul><hr /><h2 id="required-models">📥 Required Models</h2><ul><li>LTX See Motion.safetensors → <code>ComfyUI/models/checkpoints/</code> Download: <a target="_blank" rel="ugc" href="https://civarchive.com/models/2768763/ltx-see-motion?modelVersionId=3116891">https://civarchive.com/models/2768763/ltx-see-motion?modelVersionId=3116891</a></li><li>Gemma 3 12B IT FP8 text encoder (Heretic V2 used here) → <code>ComfyUI/models/text_encoders/</code> Download: <a target="_blank" rel="ugc" href="https://huggingface.co/DreamFast/gemma-3-12b-it-heretic-v2/tree/main/comfyui">https://huggingface.co/DreamFast/gemma-3-12b-it-heretic-v2/tree/main/comfyui</a></li><li>taeltx2_3.safetensors (tiny preview VAE) → <code>ComfyUI/models/vae/</code> Download: <a target="_blank" rel="ugc" href="https://huggingface.co/Kijai/LTX2.3_comfy/resolve/main/vae/taeltx2_3.safetensors?download=true">https://huggingface.co/Kijai/LTX2.3_comfy/resolve/main/vae/taeltx2_3.safetensors?download=true</a></li><li>ltx-2.3-spatial-upscaler-x2-1.1.safetensors → <code>ComfyUI/models/latent_upscale_models/</code> Download: <a target="_blank" rel="ugc" href="https://huggingface.co/Lightricks/LTX-2.3/resolve/main/ltx-2.3-spatial-upscaler-x2-1.1.safetensors?download=true">https://huggingface.co/Lightricks/LTX-2.3/resolve/main/ltx-2.3-spatial-upscaler-x2-1.1.safetensors?download=true</a></li><li>ltx-2.3-22b-distilled-lora-384-1.1.safetensors (required for FAST mode) → <code>ComfyUI/models/loras/</code> Download: <a target="_blank" rel="ugc" href="https://huggingface.co/Lightricks/LTX-2.3/resolve/main/ltx-2.3-22b-distilled-lora-384-1.1.safetensors?download=true">https://huggingface.co/Lightricks/LTX-2.3/resolve/main/ltx-2.3-22b-distilled-lora-384-1.1.safetensors?download=true</a></li></ul>⚠️ Use spatial upscaler 1.1 — version 1.0 is no longer used by this workflow. Video VAE and audio VAE load directly from the checkpoint — no separate VAE or text-projection files needed.Recommended:<ul><li>Ollama + Qwen 3.5 2B — powers the automatic reference-sheet / character analysis inside LTX Director CS. Install Ollama from <a target="_blank" rel="ugc" href="https://ollama.com">https://ollama.com</a>, then run: <code>ollama pull huihui_ai/qwen3.5-abliterated:2b</code> (1.9 GB) Without it the node still works — you just have to type the character descriptions manually.</li></ul>Optional:<ul><li>LTX-2.3 Licon MSR V2 LoRA — better multi-subject references / character sheets (keep the distilled LoRA loaded alongside): <a target="_blank" rel="ugc" href="https://huggingface.co/LiconStudio/LTX-2.3-Multiple-Subject-Reference">https://huggingface.co/LiconStudio/LTX-2.3-Multiple-Subject-Reference</a></li><li>GGUF text encoder (lower RAM; swap the loader for the GGUF-compatible LTXV Audio Text Encoder Loader): <a target="_blank" rel="ugc" href="https://huggingface.co/unsloth/gemma-3-12b-it-GGUF/tree/main">https://huggingface.co/unsloth/gemma-3-12b-it-GGUF/tree/main</a></li><li>Full official LTX-2.3 repo (base models, the distilled LoRAs and all spatial/temporal upscalers):<a target="_blank" rel="ugc" href="https://huggingface.co/Lightricks/LTX-2.3">https://huggingface.co/Lightricks/LTX-2.3</a></li></ul><hr /><h2 id="required-custom-nodes">🧩 Required Custom Nodes</h2>Install via ComfyUI-Manager (search the name) — except LTX Director CS, which must be cloned manually (see below):<ul><li>LTX Director CS (CGlide mod) — 🎛️ the timeline/director node this workflow is built on — see the red box below for installation</li><li>ComfyUI-KJNodes — SageAttention patches, LTX2 NAG, preview override — <a target="_blank" rel="ugc" href="https://github.com/kijai/ComfyUI-KJNodes">https://github.com/kijai/ComfyUI-KJNodes</a></li><li>ComfyUI-LTXVideo — tiled decode, latent-upscale loader — <a target="_blank" rel="ugc" href="https://github.com/Lightricks/ComfyUI-LTXVideo">https://github.com/Lightricks/ComfyUI-LTXVideo</a></li><li>ComfyUI-VideoHelperSuite — video combine (MP4 + audio) — <a target="_blank" rel="ugc" href="https://github.com/Kosinkadink/ComfyUI-VideoHelperSuite">https://github.com/Kosinkadink/ComfyUI-VideoHelperSuite</a></li><li>ComfyUI-Lora-Manager — LoRA loader + trigger words — <a target="_blank" rel="ugc" href="https://github.com/willmiao/ComfyUI-Lora-Manager">https://github.com/willmiao/ComfyUI-Lora-Manager</a></li><li>ComfyUI_NVIDIA_RTX_Nodes — RTX Video Super Resolution — <a target="_blank" rel="ugc" href="https://github.com/Comfy-Org/Nvidia_RTX_Nodes_ComfyUI">https://github.com/Comfy-Org/Nvidia_RTX_Nodes_ComfyUI</a></li><li>rgthree-comfy — seed control — <a target="_blank" rel="ugc" href="https://github.com/rgthree/rgthree-comfy">https://github.com/rgthree/rgthree-comfy</a></li><li>ComfyUI-Easy-Use — clean GPU VRAM — <a target="_blank" rel="ugc" href="https://github.com/yolain/ComfyUI-Easy-Use">https://github.com/yolain/ComfyUI-Easy-Use</a></li></ul>⚠️ LTX Director CS cannot be installed through the ComfyUI-Manager! Close ComfyUI, open a terminal in <code>ComfyUI/custom_nodes/</code> and run:<pre><code>git clone -b main_cs https://github.com/CGlide/WhatDreamsCost-CSGlide.git</code></pre>The <code>-b main_cs</code> part is required — the default branch does NOT contain the CS nodes (<code>LTXDirectorCS</code>, <code>LTXDirectorGuideCS</code>, …). If the Manager's "Install Missing Nodes" already created a <code>WhatDreamsCost-ComfyUI</code> folder from CGlide, delete that folder first — it's the wrong branch. The original WhatDreamsCost node is not needed for this workflow; if you have it installed for other workflows, the CS clone coexists with it.<hr /><h2 id="video-guides-walkthroughs">📺 Video Guides / Walkthroughs</h2>Step-by-step walkthroughs for every path this workflow can do. Each one is a short screen recording with on-screen chapters.<ul><li>Image to Video — the basics: turn a single image into a video with natural motion. <a target="_blank" rel="ugc" href="https://civarchive.com/images/136919778">https://civarchive.com/images/136919778</a></li><li>Video to Video — extend any existing clip to any length you like. <a target="_blank" rel="ugc" href="https://civarchive.com/images/136922039">https://civarchive.com/images/136922039</a></li><li>Audio to Video — feed a 30 s song (with vocals) + one image and get a 30 s music video, lip-synced to the track. <a target="_blank" rel="ugc" href="https://civarchive.com/images/137063968">https://civarchive.com/images/137063968</a></li><li>IC-LoRA to Video — drive motion, poses or depth from a control video using an IC-LoRA (like the one shown in the clip). <a target="_blank" rel="ugc" href="https://civarchive.com/images/137075598">https://civarchive.com/images/137075598</a></li><li>LoRA + Image to Video — multi-subject / character-sheet consistency. Needs the LTX-2.3 Licon MSR V2 LoRA. Guide: <a target="_blank" rel="ugc" href="https://civarchive.com/images/137093666">https://civarchive.com/images/137093666</a> LoRA: <code>LTX-2.3-Licon-MSR-V2.safetensors</code> Multiple-Subject-Reference LoRA by LiconStudio — keeps multiple characters and the background consistent across frames. Load it alongside the distilled LoRA.</li></ul><hr /><h2 id="installation">🛠️ Installation</h2><ol><li>Install 7 of the node packs with ComfyUI-Manager; install LTX Director CS manually with the git command above → restart ComfyUI.</li><li>Download the models into the folders listed above → restart or refresh the model list. (Optional: install Ollama + the Qwen model for automatic reference analysis.)</li><li>Open the workflow and check every red loader: checkpoint, text encoder, tiny VAE, spatial upscaler 1.1, distilled LoRA at 0.8.</li><li>Load a start image in the 🎛️ LTX Director CS, write a short motion prompt, hit Queue — the finished video lands in <code>output/</code> via the 💾 FINAL SAVE node.</li></ol><hr /><h2 id="notes">📝 Notes</h2><ul><li>⚡ RTX Video SR needs an NVIDIA RTX GPU + a recent driver. No RTX card: bypass group 8 and wire the final decode IMAGE straight into the last Video Combine.</li><li>🔮 The optional speed patches (SageAttention, Chunk FeedForward, FP16 accumulation) ship bypassed — only enable them if the <code>sageattention</code> python package is installed.</li><li>✍️ Prompting: the image defines who/what is in the scene — the prompt defines what moves. One flowing paragraph, present tense, one main action per clip. A full prompting guide is included as a note inside the workflow.</li></ul>Adult content: consenting-adult content only, where platform and law allow.Credits: Lightricks (LTX-2.3) · TenStrip (10Eros) · sulphur · WhatDreamsCost (Director) · CGlide (Director CS mod) · Kijai · checkpoint blend + workflow: SeeSee21

LTX See Motion Workflow

<pre><code>🎥 LTX SEE MOTION — IMAGE-TO-VIDEO WORKFLOW (LTX-2.3)

Two-stage ComfyUI workflow for the custom LTX See Motion LTX-2.3 checkpoint. Turns a single input image into a video with natural motion, synchronized audio and dialogue — identity, style and composition of the source image survive the whole clip.

How it works: Stage 1 renders the base clip (8 steps, CFG 1.0) → Stage 2 upscales the latent 2x and refines it (4 steps, denoise 0.42) → RTX Video SR sharpens another 1.5x → the final MP4 (H.264 + audio) saves automatically. Live low-VRAM preview included. Everything is grouped, color-coded and documented with README notes inside the workflow.

• Ships in FAST mode: distilled LoRA at strength 0.8, 8 + 4 steps, CFG 1.0
• DEV mode: disable the LoRA, set Stage 1 to 40 steps / CFG 4.0 (inside the stage subgraphs)
• Strong: image-to-video (SFW + NSFW), dancing/singing, camera moves. Weak: plain text-to-video — always feed an image
• Performance: RTX 5060 Ti 16 GB ≈ 300–400 s for a 15 s clip at ~1 MP

━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
📥 REQUIRED MODELS
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━

▪ LTX See Motion.safetensors → ComfyUI/models/checkpoints/
https://civitai.red/models/2768763/ltx-see-motion?modelVersionId=3116891

▪ Gemma 3 12B IT FP8 text encoder (Heretic V2 used here) → ComfyUI/models/text_encoders/
https://huggingface.co/DreamFast/gemma-3-12b-it-heretic-v2/tree/main/comfyui

▪ taeltx2_3.safetensors (tiny preview VAE) → ComfyUI/models/vae/
https://huggingface.co/Kijai/LTX2.3_comfy/resolve/main/vae/taeltx2_3.safetensors?download=true

▪ ltx-2.3-spatial-upscaler-x2-1.1.safetensors → ComfyUI/models/latent_upscale_models/
https://huggingface.co/Lightricks/LTX-2.3/resolve/main/ltx-2.3-spatial-upscaler-x2-1.1.safetensors?download=true

▪ ltx-2.3-22b-distilled-lora-384-1.1.safetensors (required for FAST mode) → ComfyUI/models/loras/
https://huggingface.co/Lightricks/LTX-2.3/resolve/main/ltx-2.3-22b-distilled-lora-384-1.1.safetensors?download=true

⚠️ Use spatial upscaler 1.1 — version 1.0 is no longer used by this workflow.
Video VAE and audio VAE load directly from the checkpoint — no separate VAE or text-projection files needed.

OPTIONAL:
▪ LTX-2.3 Licon MSR V2 LoRA — multi-subject references / character sheets (keep the distilled LoRA loaded alongside)
https://huggingface.co/LiconStudio/LTX-2.3-Multiple-Subject-Reference
▪ GGUF text encoder (lower RAM; swap in the GGUF-compatible LTXV Audio Text Encoder Loader)
https://huggingface.co/unsloth/gemma-3-12b-it-GGUF/tree/main
▪ Full official LTX-2.3 repo (all VAEs, LoRAs, upscalers)
https://huggingface.co/Lightricks/LTX-2.3

━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
🧩 REQUIRED CUSTOM NODES
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
Install via ComfyUI-Manager (search the name) or git clone into custom_nodes/:

• ComfyUI-KJNodes — SageAttention patches, LTX2 NAG, preview override
https://github.com/kijai/ComfyUI-KJNodes
• WhatDreamsCost-ComfyUI — LTX Director
https://github.com/WhatDreamsCost/WhatDreamsCost-ComfyUI
• ComfyUI-LTXVideo — tiled decode, latent-upscale loader
https://github.com/Lightricks/ComfyUI-LTXVideo
• ComfyUI-VideoHelperSuite — video combine (MP4 + audio)
https://github.com/Kosinkadink/ComfyUI-VideoHelperSuite
• ComfyUI-Lora-Manager — LoRA loader + trigger words
https://github.com/willmiao/ComfyUI-Lora-Manager
• ComfyUI_NVIDIA_RTX_Nodes — RTX Video Super Resolution
https://github.com/Comfy-Org/Nvidia_RTX_Nodes_ComfyUI
• rgthree-comfy — seed control
https://github.com/rgthree/rgthree-comfy
• ComfyUI-Easy-Use — clean GPU VRAM
https://github.com/yolain/ComfyUI-Easy-Use

━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
🛠️ INSTALLATION
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
1. Install the 8 node packs with ComfyUI-Manager → restart ComfyUI.
2. Download the models into the folders listed above → restart or refresh the model list.
3. Open the workflow and check every red loader: checkpoint, text encoder, tiny VAE, spatial upscaler 1.1, distilled LoRA at 0.8.
4. Load a start image in the LTX Director, write a short motion prompt, hit Queue — the finished video lands in output/ via the FINAL SAVE node.

━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
📝 NOTES
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
• RTX Video SR needs an NVIDIA RTX GPU + a recent driver. No RTX card: bypass group 8 and wire the final decode IMAGE straight into the last Video Combine.
• The optional speed patches (SageAttention, Chunk FeedForward, FP16 accumulation) ship bypassed — only enable them if the sageattention python package is installed.
• Prompting: the image defines who/what is in the scene — the prompt defines WHAT MOVES. One flowing paragraph, present tense, one main action per clip. A full prompting guide is included as a note inside the workflow.
• Adult content: consenting-adult content only, where platform and law allow.

Credits: Lightricks (LTX-2.3) · TenStrip (10Eros) · sulphur · WhatDreamsCost (Director) · Kijai · checkpoint blend + workflow: SeeSee21</code></pre>

Files with SHA256: a1f538bfbf3ce1b7e3f324aa4698892c59dc6335d141c3d211c64956da68535c