matbee comments

Results 111 comments of


                                            matbee

Updated for the `conversations` API

Looks like I'm using @jlumbroso's fork

The URL couldn't be validated. Callback verification failed with the following errors: HTTP Status Code = 404; HTTP Message = Not Found

I was getting the same error: I had a typo in my URL

The URL couldn't be validated. Callback verification failed with the following errors: HTTP Status Code = 404; HTTP Message = Not Found

The Callback url was incorrect

[META] Windows/MacOS support mega thread

// Removal of patches to the GStreamer build process in https://github.com/selkies-project/selkies-gstreamer/pull/51 means that the binary build available from the project for Windows will work without additional alterations. Does this mean...

The detected CUDA version (12.0) mismatches the version that was used to compile PyTorch (11.7)

just install 11.7 cuda and symlink /usr/local/cuda-11-7 it to /usr/local/cuda

The detected CUDA version (12.0) mismatches the version that was used to compile PyTorch (11.7)

> > just install 11.7 cuda and symlink /usr/local/cuda-11-7 it to /usr/local/cuda > > That unfortunately won't work, because my OS (Fedora 37) doesn't support gcc++ 11, only 12.. So...

Release of SD-XL unclip version

Would simply running the [SDXL-refiner](https://huggingface.co/stabilityai/stable-diffusion-xl-refiner-1.0) work as an "unclip"?

Add support for Prusa XL with up to five heads

+1 Vote for XL

How to run Ollama only on a dedicated GPU? (Instead of all GPUs)

-damn, I was not hoping for this outcome. Has anyone figured out how to restrict it to just one?- nvm, using CUDA_VISIBLE_DEVICES seemed to have done the trick

[Bug]: litellm doesnt "see" the ollama serve on wsl on Windows

> TRIED: using the hostname: host.docker.internal This fix worked on similar problem of another project... on Dialoqbase, but doesn't work on litellm 😂 I expose the host network to the...