gemma-3-12b-it-qat-q4_0-unquantized Q5_K_M

This file is exactly 8,445,037,216 bytes — 7.87 GiB / 8.45 GB at an effective 5.543 bits per weight. The nominal rate for Q5_K_M is lower; mixed-precision tensors make the real figure higher, always.

From the file· 1 file(s), summedFrom the file· 626 tensors parsed

Get it

7.87 GiB · 1 file
llama.cpp
llama-cli -hf bartowski/google_gemma-3-12b-it-qat-GGUF:Q5_K_M

Downloads and runs in one step, resolving the quantization by name.

Hugging Face CLI
hf download bartowski/google_gemma-3-12b-it-qat-GGUF google_gemma-3-12b-it-qat-Q5_K_M.gguf
Direct download

Straight from the Hugging Face CDN — we host nothing and earn nothing from this. Verify what you received against the exact byte count above; a size mismatch is the usual cause of a file that will not load.

Size
7.87 GiB
8.45 GB
Effective bpw
5.543
from real bytes ÷ params
Tensors
626
Header
6.55 MB
GGUF metadata

What this quantization actually contains

per-tensor types, parsed from the GGUF header
F32
289
Q5_K
288
Q6_K
49

A quantization label names a mixture, not a uniform precision. Attention and output tensors are routinely kept at higher precision than the label implies, which is exactly why the effective bits-per-weight above exceeds the nominal rate.