[sample] seed=2027 640x480 frames=81 steps=8 split=3 [32m[INFO][0m got prompt [comfy] executing vae [32m[INFO][0m Using pytorch attention in VAE [32m[INFO][0m Using pytorch attention in VAE [32m[INFO][0m VAE load device: cuda:0, offload device: cpu, dtype: torch.bfloat16 [comfy] executing image [comfy] executing vision [32m[INFO][0m Requested to load CLIPVisionModelProjection [comfy] executing features [32m[INFO][0m Model CLIPVisionModelProjection prepared for dynamic VRAM loading. 1205MB Staged. 0 patches attached. Force pre-loaded 132 weights: 330 KB. [comfy] executing clip [32m[INFO][0m Found quantization metadata version 1 [32m[INFO][0m Using MixedPrecisionOps for text encoder [32m[INFO][0m CLIP/text encoder model load device: cuda:0, offload device: cpu, current: cpu, dtype: torch.float16 [32m[INFO][0m Requested to load WanTEModel [comfy] executing negative [32m[INFO][0m Model WanTEModel prepared for dynamic VRAM loading. 6419MB Staged. 0 patches attached. Force pre-loaded 73 weights: 488 KB. [comfy] executing positive [32m[INFO][0m Model WanTEModel prepared for dynamic VRAM loading. 6419MB Staged. 0 patches attached. Force pre-loaded 73 weights: 488 KB. [comfy] executing conditioning [32m[INFO][0m Requested to load WanVAE [32m[INFO][0m Model WanVAE prepared for dynamic VRAM loading. 241MB Staged. 0 patches attached. Force pre-loaded 60 weights: 61 KB. [comfy] executing high [32m[INFO][0m Found quantization metadata version 1 [32m[INFO][0m Detected mixed precision quantization [32m[INFO][0m Using mixed precision operations [32m[INFO][0m Native ops: float8_e4m3fn, convrot_w4a4, int8_tensorwise, float8_e5m2, asym_w4a8_int8 , emulated ops: mxfp8, nvfp4 [32m[INFO][0m model weight dtype torch.float16, manual cast: torch.float16 [32m[INFO][0m model_type FLOW [comfy] executing high_shift [comfy] executing sample_high [32m[INFO][0m Requested to load WAN21 [32m[INFO][0m Model WAN21 prepared for dynamic VRAM loading. 13853MB Staged. 0 patches attached. 0%| | 0/3 [00:00