Video generation models are evolving quickly, but many of them require large GPU memory and complicated setups. The LTX-2.3 model is a lightweight alternative that can run efficiently using GGUF quantized models.
If you’re thinking about purchasing a new GPU, we’d greatly appreciate it if you used our Amazon Associate links. The price you pay will be exactly the same, but Amazon provides us with a small commission for each purchase. It’s a simple way to support our site and helps us keep creating useful content for you. Recommended GPUs: RTX 5090, RTX 5080, and RTX 5070. #ad
With the GGUF format, LTX-2.3 becomes much easier to run locally, even on GPUs with limited VRAM. When combined with ComfyUI, you can build flexible workflows that generate videos from either text prompts or input images.
In this guide, you will learn how to run LTX-2.3 GGUF in ComfyUI for image-to-video generation. The same workflow can also be used for text-to-video, allowing you to create short AI videos directly from prompts.
We will walk through the required models, how to load the workflow, and the basic settings needed to generate your first video.
LTX-2.3 Models
- GGUF Model: The GGUF models can be found here. I have a RTX 5090, and I used the Q8 variant. I downloaded ltx-2.3-22b-dev-Q8_0.gguf. If you have less VRAM, use other variants like Q6 or Q4. Put the GGUF models in ComfyUI\models\unet\ .
- Text Encoders: Download gemma_3_12B_it.safetensors and ltx-2.3_text_projection_bf16.safetensors; put them in ComfyUI\models\text_encoders\ .
- VAE: Download both the video VAE and audio VAE from LTX-2.3 VAE Folder, put them in ComfyUI\models\vae\ .
- Latent Upscale Mode: Download ltx-2.3-spatial-upscaler-x2-1.0.safetensors , and put it under ComfyUI\models\latent_upscale_models\ .
- LoRA: Download ltx-2.3-22b-distilled-lora-384.safetensors . Place them in ComfyUI\models\loras\ .
LTX-2.3 Image-to-Video GGUF Installation
- Update your ComfyUI to the latest version if you haven’t already. (Run update\update_comfyui.bat for Windows). You also need to update ComfyUI-GGUF, ComfyUI-LTXVideo, and ComfyUI-KJNodes to the latest versions.
- Download the json file, and open it using ComfyUI.
- Use ComfyUI Manager to install missing nodes.
- Restart ComfyUI.
LTX-2.3 Workflow Nodes
Load Models Group
Select the GGUF model you downloaded.
Choose the video VAE and the audio VAE.
Specify the latent upscale model.
Pick the distilled LoRA.
Load Image Group
Choose a file if you are using image-to-video
Set this to true if you want to use text-to-video
Set Prompt Group
Pick the two text encoders.
Enter the postive prompt and the negative prompt here.
Set the frames per second.
Number of frames to be generated. For 24fps, this is about 10 seconds.
LTX-2.3 Image-to-Video Examples
Input image:
Prompt:
The woman gently tilts the can and takes a refreshing sip, her eyes closing slightly with pleasure. A light breeze makes her hair and dress flutter The camera slowly pans and tilts upward as the sunlight flares more intensely behind her, creating a dreamy golden shimmer. Light lens flares, soft wind movement in the wheat field, subtle camera shake for realism. Warm and radiant motion, smooth transitions, soft glowing ambiance, cinematic light bloom, 16:9 aspect ratio.
LTX-2.3:
LTX-2:
Wan 2.2:
Input image:
Prompt:
The camera tracks the sleek black sports car as it races down a wet, neon-lit city street at night. Reflections of magenta, cyan, and red lights shimmer on the car’s glossy surface and the wet asphalt. The car accelerates slightly as the lights streak past in the background, with a subtle motion blur and tire spray. Its headlights flare and cast sharp beams forward, illuminating the wet road ahead. The camera rotates around the front-left side of the car, highlighting its curves and aggressive stance. Soft raindrops hit the windshield in slow motion. Soundless, but with cinematic tension. High contrast lighting, futuristic tone, slow motion elements, hyper-realistic motion, 16:9 aspect ratio
LTX-2.3:
LTX-2:
Wan 2.2:
Input image:
Prompt:
The woman runs steadily forward, her steps rhythmic and powerful. Her ponytail bounces with each stride as warm morning light ripples across her body and the bridge. Subtle camera shake adds realism as the scene follows her from a side angle. A light breeze moves her clothing naturally. The sun rises behind her, casting golden flares through the bridge cables. Drops of sweat glisten and roll down her skin in slow motion. The video closes with her stopping to catch her breath, turning toward the camera with a confident smile. Realistic motion, slow-to-normal pacing blend, dynamic light transitions, motivational mood, cinematic tone, 16:9 format
LTX-2.3:
LTX-2:
Wan 2.2:
Input image:
Prompt:
The woman slowly lifts and puts on her sunglasses as the golden sun sets behind her. Her hair moves gently in the wind, and the reflection in the lenses captures the glowing city skyline. As the glasses settle on her face, the light subtly shifts, casting a cinematic flare across the lens. The camera slowly pushes in toward her face, enhancing the cool, composed mood. Lens flares, soft camera movement, golden hour light, confident tone, 16:9 aspect ratio
LTX-2.3:
LTX-2:
Wan 2.2:
Input image:
Prompt:
A young woman sits in a cozy modern café, facing the camera at eye level. She smiles gently and speaks directly to the viewer in a calm, friendly tone. Her lips sync naturally as she says: “It’s kind of amazing… with the release of LTX-2, I can finally talk to you like this. It feels more real, more alive. If you want to see what I create next, follow me and stay with me.” Her facial expressions are subtle and natural, with soft eye contact, slight head movements, and small hand gestures near a coffee cup on the table. The motion is smooth and coherent, with stable facial structure and consistent identity throughout the clip. The café background remains steady and realistic, with minimal camera movement, no exaggerated motion, and no stylization. Natural daylight illuminates her face evenly, maintaining photorealistic skin texture, realistic lip movement, and believable human timing. The overall mood is warm, intimate, and conversational, as if she is casually talking to the viewer in real life.
LTX-2.3:
LTX-2:
No equivalent Wan 2.2 output with dialog.
LTX-2.3 Test-to-Video Examples
COMING SOON…
Conclusion
LTX-2.3 GGUF makes local AI video generation much more accessible. By using quantized models, you can run image-to-video and text-to-video workflows in ComfyUI without the heavy hardware requirements of many newer video models.
Once the workflow is set up, generating videos becomes straightforward. You can experiment with different prompts, starting images, frame counts, and motion settings to produce a wide variety of results.
Because ComfyUI is modular, it is also easy to expand this workflow. You can add upscaling, frame interpolation, or other post-processing steps to improve video quality and smoothness.
If you are interested in local AI video generation, LTX-2.3 GGUF in ComfyUI is a practical and efficient place to start.
Leave a Reply