gemma-7b-exl2

Exllama v2 Quantizations of gemma-7b

Using <a href="https://github.com/turboderp/exllamav2/releases/tag/v0.0.13">turboderp's ExLlamaV2 v0.0.13</a> for quantization.

<b>The "main" branch only contains the measurement.json, download one of the other branches for the model (see below)</b>

Each branch contains an individual bits per weight, with the main one containing only the meaurement.json for further conversions.

Original model: https://huggingface.co/google/gemma-7b

No GQA - VRAM requirements will be higher

Download instructions

With git:

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

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

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

Linux:

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

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

mkdir gemma-7b-exl2-6.5
huggingface-cli download bartowski/gemma-7b-exl2 --revision 6_5 --local-dir gemma-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