Files
runpod/comfyui
Sebastian Krüger 385b36b062 feat: enable CPU offload for CogVideoX model to reduce VRAM usage
- Add enable_sequential_cpu_offload=true to DownloadAndLoadCogVideoModel
- Reduces VRAM from ~20GB to ~12GB at cost of slower inference
- Widget values: [model, precision, quantization, cpu_offload] = ['THUDM/CogVideoX-5b-I2V', 'bf16', 'disabled', true]
- Necessary for 24GB GPU with other services running
2025-11-23 09:47:02 +01:00
..