austinrick

v0.3.76 is noticeably much slower than v0.3.75 when running(z-image model) KSampler. Please refer to the log below： [comfyui.log_2025-12-01T17-28-53-130Z ComfyUI version 0.3.75.log](https://github.com/user-attachments/files/23881307/comfyui.log_2025-12-01T17-28-53-130Z.ComfyUI.version.0.3.75.log) [comfyui.log_2025-12-02T08-16-46-911Z ComfyUI version 0.3.76.log](https://github.com/user-attachments/files/23881308/comfyui.log_2025-12-02T08-16-46-911Z.ComfyUI.version.0.3.76.log)

> v0.3.76 is noticeably much slower than v0.3.75 when running(z-image model) KSampler. Please refer to the log below： [comfyui.log_2025-12-01T17-28-53-130Z ComfyUI version 0.3.75.log](https://github.com/user-attachments/files/23881307/comfyui.log_2025-12-01T17-28-53-130Z.ComfyUI.version.0.3.75.log) [comfyui.log_2025-12-02T08-16-46-911Z ComfyUI version 0.3.76.log](https://github.com/user-attachments/files/23881308/comfyui.log_2025-12-02T08-16-46-911Z.ComfyUI.version.0.3.76.log) Hi All, After testing...

> Are you 100% sure you're using the exact same workflow setup in both tests? If you used a different sampler, it could slow things down. Like if you used...

> Can you post screenshots of your while inferencing, for both versions? You're already offloading in both cases, but it's possible that one of them is using more shared memory...