QAT version

#3
by Delnith - opened

Is the team planning on utilizing the recently released QAT versions for a 4-bit quantization?

https://huggingface.co/google/gemma-3-27b-it-qat-q4_0-unquantized

From their press releases, this apparently was trained in such a way as to make it almost identical when quantized.

Sign up or log in to comment