Please, release 8B - 12B version
#9
by AlxHQ - opened
It would be nice to be able to run the model on a single 16GB GPU, but the current 26B and 31B models don't fit into a 16GB VRAM, and the lower-end models are much dumber. So why isn't there a middle option?
Hi @AlxHQ
Check this out https://blog.google/innovation-and-ai/technology/developers-tools/introducing-gemma-4-12b/
Feel free to share the feedback . Thanks