LTXV2.3 GGUF I2V/T2V/FFLF+IDaudioLora 12gb Vram sageattn — v1.0

Author notes from video_ltx2_i2v.json · selected release

## LTX-2 Prompting Tips 1. **Core Actions**: Describe events and actions as they occur over time 2. **Visual Details**: Describe all visual details you want to appear in the video 3. **Audio**: Describe sounds and dialogue needed for the scene ## Report LTX-2 Issues To report any issues when running this workflow, go to: Coming soon --- ## Model links ### **Diffusion Models (GGUF – from Kijai)** **Choose ONE dev model and/or ONE distilled model** - [ltx-2-19b-dev_Q4_K_M.gguf](https://huggingface.co/Kijai/LTXV2_comfy/resolve/main/diffusion_models/ltx-2-19b-dev_Q4_K_M.gguf) - [ltx-2-19b-dev_Q8_0.gguf](https://huggingface.co/Kijai/LTXV2_comfy/resolve/main/diffusion_models/ltx-2-19b-dev_Q8_0.gguf) - [ltx-2-19b-distilled_Q4_K_M.gguf](https://huggingface.co/Kijai/LTXV2_comfy/resolve/main/diffusion_models/ltx-2-19b-distilled_Q4_K_M.gguf) - [ltx-2-19b-distilled_Q6_K.gguf](https://huggingface.co/Kijai/LTXV2_comfy/resolve/main/diffusion_models/ltx-2-19b-distilled_Q6_K.gguf) - [ltx-2-19b-distilled_Q8_0.gguf](https://huggingface.co/Kijai/LTXV2_comfy/resolve/main/diffusion_models/ltx-2-19b-distilled_Q8_0.gguf) --- ### **Text Encoders** **Gemma 3 (official, unchanged)** - [gemma_3_12B_it.safetensors](https://huggingface.co/Comfy-Org/ltx-2/resolve/main/split_files/text_encoders/gemma_3_12B_it.safetensors) **LTX-2 embedding connectors (from Kijai)** - [ltx-2-19b-embeddings_connector_dev_bf16.safetensors](https://huggingface.co/Kijai/LTXV2_comfy/resolve/main/text_encoders/ltx-2-19b-embeddings_connector_dev_bf16.safetensors) - [ltx-2-19b-embeddings_connector_distill_bf16.safetensors](https://huggingface.co/Kijai/LTXV2_comfy/resolve/main/text_encoders/ltx-2-19b-embeddings_connector_distill_bf16.safetensors) --- ### **VAE Models (from Kijai)** - [LTX2_video_vae_bf16.safetensors](https://huggingface.co/Kijai/LTXV2_comfy/resolve/main/VAE/LTX2_video_vae_bf16.safetensors) - [LTX2_audio_vae_bf16.safetensors](https://huggingface.co/Kijai/LTXV2_comfy/resolve/main/VAE/LTX2_audio_vae_bf16.safetensors) --- ### **LoRAs** **Distilled LoRA (from Kijai)** - [ltx-2-19b-distilled-lora-384.safetensors](https://huggingface.co/Kijai/LTXV2_comfy/resolve/main/loras/ltx-2-19b-distilled-lora-384.safetensors) **Camera Control (official)** - [ltx-2-19b-lora-camera-control-dolly-left.safetensors](https://huggingface.co/Lightricks/LTX-2-19b-LoRA-Camera-Control-Dolly-Left/resolve/main/ltx-2-19b-lora-camera-control-dolly-left.safetensors) --- ### **Latent Upscale Models (official)** - [ltx-2-spatial-upscaler-x2-1.0.safetensors](https://huggingface.co/Lightricks/LTX-2/resolve/main/ltx-2-spatial-upscaler-x2-1.0.safetensors)

Author notes from video_ltx2_i2v.json · selected release

Width & height settings must be divisible by 32 + 1. Frame count must be divisible by 8 + 1. Running with invalid parameters **will not cause errors**. Instead, the flow will silently choose the closest valid parameters. By default, we are using 720p resolution. You can try 1920*1088 if you have a powerful GPU.

Author notes from video_ltx2_i2v.json · selected release

## - [ComfyUI-GGUF](https://github.com/city96/ComfyUI-GGUF) — GGUF Node pack required ## - [ComfyUI-KJNodes](https://github.com/kijai/ComfyUI-KJNodes) — Install or update this node pack to get the Audio VAE node #If you are getting checkerboard patterned noise output Updated all nodes and ComfyUI itself!!

Author notes from video_ltx2_i2v.json · selected release

Please make sure the frame rate value is the same as the Frame rate float.

Author notes from video_ltx2_i2v.json · selected release

## Settings: Distilled 8steps cfg 1 Dev 50 cfg 4