Pull the image
Pre-built and pushed to Docker Hub. No local build needed.
docker pull timpietruskyrunpod/comfyui-wizard:tpl-ltxv_text_to_video
Dockerfile
# clean base image containing only comfyui, comfy-cli and comfyui-manager
FROM runpod/worker-comfyui:5.8.4-base
# download models into comfyui
RUN comfy model download --url https://huggingface.co/comfyanonymous/flux_text_encoders/resolve/main/t5xxl_fp16.safetensors?download=true --relative-path models/text_encoders --filename t5xxl_fp16.safetensors
RUN comfy model download --url https://huggingface.co/Lightricks/LTX-Video/resolve/main/ltx-video-2b-v0.9.safetensors?download=true --relative-path models/checkpoints --filename ltx-video-2b-v0.9.safetensors
Run Instructions
Build
docker build -t comfyui-tpl-ltxv_text_to_video .
Run
docker run --gpus all -p 8188:8188 timpietruskyrunpod/comfyui-wizard:tpl-ltxv_text_to_video
RunPod Serverless
Use the Dockerfile as a RunPod Serverless worker template