Orso Eric

Same issue trying to install oobabooga

It works like a charm. Thanks for this code slmisc! I was working on building a DMA capable LCD library for the LonganNano GD32VF103 as well, this will save lots...

I'm from the future of 2023. The seek/focus functions in this thread have been useful telling the solver what I need, I was having an hard time getting it to...

> tested with triton==3.2.0, and it seems to work, not sure if pinning triton version in requirements.txt is a good idea though I had the same issue, and your proposed...

> pip3 install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/rocm6.3 I tried but it bricked ComfyUI, I'm trying to bring it back > Versions of relevant libraries: > [pip3] alias-free-torch==0.0.6 > [pip3]...

Yeah, I'm rebuilding WSL a different way now that I know more about it. I built it like this because it was my tenth attempt to get ROCm to run,...

So, I rebuilt WSL with a number of changes (https://github.com/OrsoEric/HOWTO-7900XTX-Win-ROCM?tab=readme-ov-file#step-1---wsl) I'm using UV as virtual environment and did some changes to the AMD instruction to get it all local and...

> So it's working on rocm 6.3? It does not work, it's the same issue I have with ROCm 6.3.4. With ROcm 6.3 I get driver timeout on VAE decode...

I'll try friday, I'll be busy diffusing today and I need it to work. Replicating the OOM on VAE decode is really easy. I have had a regression in performance...

> I have had a regression in performance with the python 3.12 binaries. I'll also try with a VENV with the python 3.10 binaries. Hidream takes around 260s which is...