Run Mixtral-8x7B models in Colab or consumer desktops
Why do you think that https://github.com/stochasticai/xTuring is a good alternative to mixtral-offloading
Run Mixtral-8x7B models in Colab or consumer desktops
Why do you think that https://github.com/stochasticai/xTuring is a good alternative to mixtral-offloading