Bingsu_-_llama-190m-arch-exl2

Quantization made by Richard Erkhov.

Github

Discord

Request more models

llama-190m-arch - EXL2

• Model creator: https://huggingface.co/Bingsu/
• Original model: https://huggingface.co/Bingsu/llama-190m-arch/

Available sizes

| Branch | Bits | Description | | ----- | ---- | ------- | ------ | ------ | ------ | ------ | ------------ | | 8_0 | 8.0 | Maximum quality that ExLlamaV2 can produce, near unquantized performance. | | 6_5 | 6.5 | Very similar to 8.0, good tradeoff of size vs performance, recommended. | | 5_0 | 5.0 | Slightly lower quality vs 6.5, but usable on 8GB cards. | | 4_25 | 4.25 | GPTQ equivalent bits per weight, slightly higher quality. | | 3_5 | 3.5 | Lower quality, only use if you have to. |

Download instructions

With git:

git clone --single-branch --branch 6_5 https://huggingface.co/Bingsu_-_llama-190m-arch-exl2 llama-190m-arch-6_5

With huggingface hub:

pip3 install huggingface-hub

To download a specific branch, use the --revision parameter. For example, to download the 6.5 bpw branch: Linux:

huggingface-cli download Bingsu_-_llama-190m-arch-exl2 --revision 6_5 --local-dir llama-190m-arch-6_5 --local-dir-use-symlinks False

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

huggingface-cli download Bingsu_-_llama-190m-arch-exl2 --revision 6_5 --local-dir llama-190m-arch-6.5 --local-dir-use-symlinks False

Original model description: Entry not found