metadata
license: apache-2.0
inference: false
quantized_by: bartowski
Exllama v2 Quantizations of MistralLite at 8.0,6.0,4.0
Using turboderp's ExLlamaV2 v0.0.7 for quantization.
Conversion was done using wikitext-103-raw-v1-test.parquet as calibration dataset.
Original model: https://huggingface.co/amazon/MistralLite
Download instructions
With git:
git clone --single-branch --branch 8.0,6.0,4.0 https://huggingface.co/bartowski/MistralLite-exl2
With huggingface hub (credit to TheBloke for instructions):
pip3 install huggingface-hub
To download from a different branch, add the --revision parameter:
mkdir MistralLite-exl2
huggingface-cli download bartowski/MistralLite-exl2 --revision --local-dir MistralLite-exl2 --local-dir-use-symlinks False