Exllama v2 Quantizations of Qwen1.5-32B-Chat
Using turboderp's ExLlamaV2 v0.0.17 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/Qwen/Qwen1.5-32B-Chat
Branch | Bits | lm_head bits | VRAM (4k) | VRAM (16k) | VRAM (32k) | Description |
---|---|---|---|---|---|---|
6_5 | 6.5 | 8.0 | 28.9 GB | 31.6 GB | 35.6 GB | Near unquantized performance at vastly reduced size, recommended. |
4_25 | 4.25 | 6.0 | 19.5 GB | 22.2 GB | 26.2 GB | GPTQ equivalent bits per weight, slightly higher quality. |
3_5 | 3.5 | 6.0 | 16.5 GB | 19.2 GB | 23.2 GB | Lower quality, only use if you have to. |
3_0 | 3.0 | 6.0 | 14.3 GB | 17.0 GB | 21.0 GB | Very low quality, usable with 16gb of VRAM. |
Download instructions
With git:
git clone --single-branch --branch 6_5 https://huggingface.co/bartowski/Qwen1.5-32B-Chat-exl2 Qwen1.5-32B-Chat-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 Qwen1.5-32B-Chat-exl2
:
mkdir Qwen1.5-32B-Chat-exl2
huggingface-cli download bartowski/Qwen1.5-32B-Chat-exl2 --local-dir Qwen1.5-32B-Chat-exl2 --local-dir-use-symlinks False
To download from a different branch, add the --revision
parameter:
Linux:
mkdir Qwen1.5-32B-Chat-exl2-6_5
huggingface-cli download bartowski/Qwen1.5-32B-Chat-exl2 --revision 6_5 --local-dir Qwen1.5-32B-Chat-exl2-6_5 --local-dir-use-symlinks False
Windows (which apparently doesn't like _ in folders sometimes?):
mkdir Qwen1.5-32B-Chat-exl2-6.5
huggingface-cli download bartowski/Qwen1.5-32B-Chat-exl2 --revision 6_5 --local-dir Qwen1.5-32B-Chat-exl2-6.5 --local-dir-use-symlinks False
Want to support my work? Visit my ko-fi page here: https://ko-fi.com/bartowski