-
Hey i am getting this issue while running flux forge on MAC OS 15.0.1 M3 pro apple silicone
-
### Your current environment
pip3 install vllm==0.4.2 nvidia-ammo==0.7.1
Collecting environment information...
PyTorch version: 2.3.0+cu121
Is debug build: False
CUDA used to build PyTorch: …
-
## ❓ Question
I'm trying to run the `examples/dynamo/vgg16_fp8_ptq.y` example but got following error:
```
Traceback (most recent call last):
File "/home/wh/generative_action/SynHSI/vgg_quat.p…
-
https://pytorch.org/docs/stable/notes/amp_examples.html
Currently, `bfloat16` works well without grad scaling. But to use `fp16` and `fp8` (`fp8` - in the future, when the support for Hopper/40XX G…
-
### Expected Behavior
To render the image
### Actual Behavior
no image it bugs
### Steps to Reproduce
Using the ComfyUI workflow present in wiki page
### Debug Logs
```powershell
I can not pu…
-
As a Stable Diffusion user, new torch-directml is needed for issue listed here:
bug issue:
1. bad memory allocation: GPU memory leak after every generation; [https://github.com/vladmandic/automati…
-
### Expected Behavior
I'm having a heck of a time finding a working Torch to just work ... I dunno what happened, but I upraded (all) and it borked my install.. now when I try a comy lora/flux workfl…
-
First of all, Thank you for sharing your excellent work!
I have a question about overlapping (pingpong design). From my understanding:
1) With FP8 precision and a head dimension of 128, the expo…
-
Hi, I noticed that the FP8LinearStatic will dequantize the output(fp8) to input dtype(fp16/bf16) as below L209 shows.
1. Is it due to the attention kernel does not support fp8 at that time? And is it…
-
Hi @hongxiayang @hliuca ,
It seems like float8 training using `torchao.float8` is not support at the moment. Is there a different library or code path I should be using for float8 training or what …