You can deploy this model and start earning money today!

bartowski/
cosmo-1b-exl2

Name: cosmo-1b-exl2
Availability: InStock

text-generation

HuggingFaceTB/cosmopedia

apache-2.0

Views

No views yet

Exllama v2 Quantizations of cosmo-1b

Using turboderp's ExLlamaV2 v0.0.13 for quantization.

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

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

Conversion was done using the default calibration dataset.

Default arguments used except when the bits per weight is above 6.0, at that point the lm_head layer is quantized at 8 bits per weight instead of the default 6.

Original model: https://huggingface.co/HuggingFaceTB/cosmo-1b

Download instructions

With git:

git clone --single-branch --branch 6_5 https://huggingface.co/bartowski/cosmo-1b-exl2

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 cosmo-1b-exl2:

shell

mkdir cosmo-1b-exl2
huggingface-cli download bartowski/cosmo-1b-exl2 --local-dir cosmo-1b-exl2 --local-dir-use-symlinks False

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

Linux:

shell

mkdir cosmo-1b-exl2-6_5
huggingface-cli download bartowski/cosmo-1b-exl2 --revision 6_5 --local-dir cosmo-1b-exl2-6_5 --local-dir-use-symlinks False

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

shell

mkdir cosmo-1b-exl2-6.5
huggingface-cli download bartowski/cosmo-1b-exl2 --revision 6_5 --local-dir cosmo-1b-exl2-6.5 --local-dir-use-symlinks False