image generation extremy slow with RTX 5090

Basically my images with the same format are generated in 15 to 18 seconds, now it takes 40 seconds, a disgrace, same GPU as always, RTX 5090 POD ID: 2pbfz0u9uk836f - Restoring initial comfy attention Patching torch settings: torch.backends.cuda.matmul.allow_fp16_accumulation = False Requested to load WanVAE loaded completely 294.1602478027344 242.02829551696777 True Prompt executed in 39.48 seconds Requested to load WAN21 loaded completely 30086.92495803833 27251.406372070312 True Patching comfy attention to use sageattn Patching torch settings: torch.backends.cuda.matmul.allow_fp16_accumulation = True (RES4LYF) rk_type: res_2s 100%|██████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 4/4 [00:03<00:00, 1.02it/s] Restoring initial comfy attention Patching torch settings: torch.backends.cuda.matmul.allow_fp16_accumulation = False Requested to load WAN21 loaded completely 30086.92495803833 27251.406372070312 True Patching comfy attention to use sageattn Patching torch settings: torch.backends.cuda.matmul.allow_fp16_accumulation = True (RES4LYF) rk_type: res_2s 100%|██████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 6/6 [00:05<00:00, 1.01it/s] Restoring initial comfy attention Patching torch settings: torch.backends.cuda.matmul.allow_fp16_accumulation = False Requested to load WanVAE loaded completely 294.1602478027344 242.02829551696777 True Prompt executed in 39.04 seconds Requested to load WAN21
0 Replies
No replies yetBe the first to reply to this messageJoin

Did you find this page helpful?