Branch	Bits	lm_head bits	Size (4k)	Size (16k)	Description
8_0	8.0	8.0	14.0 GB	19.4 GB	Maximum quality that ExLlamaV2 can produce, near unquantized performance.
6_5	6.5	8.0	12.5 GB	17.9 GB	Near unquantized performance at vastly reduced size, recommended.
5_0	5.0	6.0	10.9 GB	16.3 GB	Slightly lower quality vs 6.5, but usable on 8GB cards with 4k context.
4_25	4.25	6.0	10.2 GB	15.7 GB	GPTQ equivalent bits per weight.
3_5	3.5	6.0	9.5 GB	14.9 GB	Lower quality, not recommended.

Download instructions

With git:

git clone --single-branch --branch 6_5 https://huggingface.co/bartowski/codegemma-7b-exl2 codegemma-7b-exl2-6_5

With huggingface hub (credit to TheBloke for instructions):

pip3 install huggingface-hub

To download the main (only useful if you only care about measurement.json) branch to a folder called codegemma-7b-exl2:

shell

mkdir codegemma-7b-exl2
huggingface-cli download bartowski/codegemma-7b-exl2 --local-dir codegemma-7b-exl2 --local-dir-use-symlinks False

To download from a different branch, add the --revision parameter:

Linux:

shell

mkdir codegemma-7b-exl2-6_5
huggingface-cli download bartowski/codegemma-7b-exl2 --revision 6_5 --local-dir codegemma-7b-exl2-6_5 --local-dir-use-symlinks False

Windows (which apparently doesn't like _ in folders sometimes?):

shell

mkdir codegemma-7b-exl2-6.5
huggingface-cli download bartowski/codegemma-7b-exl2 --revision 6_5 --local-dir codegemma-7b-exl2-6.5 --local-dir-use-symlinks False

Want to support my work? Visit my ko-fi page here: https://ko-fi.com/bartowski