LTX2 GGUF — Low VRAM Video Generation (I2V / T2V)
By: The_frizzy1 Hardware target: RTX 3050 (4 GB) / RTX 3060 (12 GB) CivitAI: https://civitai.com/models/2339823/ltx2-gguf-low-vram-video-generation-i2v-t2v YouTube: https://www.youtube.com/@the_frizzy1
🎥 Video Explainer: https://www.youtube.com/watch?v=nnHUBMgdJac
What This Is
Simple color-coded GGUF workflow for LTXV-2 in ComfyUI. Lightweight, I2V and T2V support.
Tested Hardware
| Hardware | Notes |
|---|---|
| RTX 3060 12GB + 48GB RAM | Main test hardware |
| RTX 3050 4GB + 16GB RAM | Enable Triton |
Common Issues
LTXVEmptyLatentAudio – AttributeError→ Update KJNodes to v1.2.8+- Checkerboard noise → Update all nodes + ComfyUI
Required Custom Nodes
| Node | Link |
|---|---|
| ComfyUI-GGUF | https://github.com/city96/ComfyUI-GGUF |
| KJNodes (keep updated!) | https://github.com/kijai/ComfyUI-KJNodes |
Model Downloads
Main: https://huggingface.co/Kijai/LTXV2_comfy/tree/main/
Diffusion (pick one):
Text Encoders (both required):
- https://huggingface.co/Comfy-Org/ltx-2/tree/main/split_files/text_encoders
- https://huggingface.co/Kijai/LTXV2_comfy/resolve/main/text_encoders/ltx-2-19b-embeddings_connector_dev_bf16.safetensors
VAE (both files): https://huggingface.co/Kijai/LTXV2_comfy/tree/main/VAE
Upscaler: https://huggingface.co/Lightricks/LTX-2/tree/main/ltx-2-spatial-upscaler-x2-1.0.safetensors
LoRAs:
Changelog
| Version | Notes |
|---|---|
| v1.5 | Current release |
| v1.0 | Initial release |