bartowski's picture
Update README.md
1f5ce47 verified
metadata
license: mit
license_link: https://huggingface.co/microsoft/Phi-3-medium-4k-instruct/resolve/main/LICENSE
language:
  - multilingual
pipeline_tag: text-generation
tags:
  - nlp
  - code
inference:
  parameters:
    temperature: 0.7
widget:
  - messages:
      - role: user
        content: Can you provide ways to eat combinations of bananas and dragonfruits?
quantized_by: bartowski

Exllama v2 Quantizations of Phi-3-medium-4k-instruct

Using turboderp's ExLlamaV2 v0.0.21 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.

Original model: https://huggingface.co/microsoft/Phi-3-medium-4k-instruct

Prompt format

<|user|>
{prompt}<|end|>
<|assistant|>

Available sizes

Branch Bits lm_head bits VRAM (4k) VRAM (16k) Description
8_0 8.0 8.0 14.0 GB 16.4 GB Max quality that ExLlamaV2 can produce, recommended.
6_5 6.5 8.0 12.5 GB 14.9 GB Near unquantized performance at vastly reduced size, recommended.
5_0 5.0 6.0 10.0 GB 12.4 GB Slightly lower quality vs 6.5.
4_25 4.25 6.0 8.8 GB 11.2 GB GPTQ equivalent bits per weight.
3_5 3.5 6.0 7.6 GB 10.0 GB Lower quality, not recommended.

Download instructions

With git:

git clone --single-branch --branch 6_5 https://huggingface.co/bartowski/Phi-3-medium-4k-instruct-exl2 Phi-3-medium-4k-instruct-exl2-6_5

With huggingface hub (credit to TheBloke for instructions):

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 bartowski/Phi-3-medium-4k-instruct-exl2 --revision 6_5 --local-dir Phi-3-medium-4k-instruct-exl2-6_5

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

huggingface-cli download bartowski/Phi-3-medium-4k-instruct-exl2 --revision 6_5 --local-dir Phi-3-medium-4k-instruct-exl2-6.5

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