Richard

+1

Although I had nothing to do with the creation of this project, I see that it is using [llama.cpp](https://github.com/ggerganov/llama.cpp), a project that focuses on using CPU vectorization to run the...

@AntouanK @farrael004 you guys can take a look at this issue where they discuss inference in consumer-grade GPUs https://github.com/facebookresearch/llama/issues/4

This appears to be a reasonable request. Requiring a newly created user to log in is detrimental to user experience. Hope this gets solved 👍 .