Exllama v2 Quantizations of wavecoder-ultra-6.7b
Using turboderp's ExLlamaV2 v0.0.18 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/wavecoder-ultra-6.7b
Prompt format
This seems to follow the DeepSeek coder format, aka Alpaca.
{system_prompt}
### Instruction: {prompt}
### Response:
Available sizes
No GQA - VRAM requirements will be higher
| Branch | Bits | lm_head bits | Size (4k) | Size (16k) | Description |
|---|---|---|---|---|---|
| 8_0 | 8.0 | 8.0 | 9.0 GB | 15.2 GB | Maximum quality that ExLlamaV2 can produce, near unquantized performance. |
| 6_5 | 6.5 | 8.0 | 8.2 GB | 14.4 GB | Near unquantized performance at vastly reduced size, recommended. |
| 5_0 | 5.0 | 6.0 | 6.8 GB | 13.0 GB | Slightly lower quality vs 6.5, but usable on 8GB cards with 4k context. |
| 4_25 | 4.25 | 6.0 | 6.1 GB | 12.3 GB | GPTQ equivalent bits per weight. |
| 3_5 | 3.5 | 6.0 | 5.5 GB | 11.7 GB | Lower quality, not recommended. |
Download instructions
With git:
git clone --single-branch --branch 6_5 https://huggingface.co/bartowski/wavecoder-ultra-6.7b-exl2 wavecoder-ultra-6.7b-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/wavecoder-ultra-6.7b-exl2 --revision 6_5 --local-dir wavecoder-ultra-6.7b-exl2-6_5 --local-dir-use-symlinks False
Windows (which apparently doesn't like _ in folders sometimes?):
huggingface-cli download bartowski/wavecoder-ultra-6.7b-exl2 --revision 6_5 --local-dir wavecoder-ultra-6.7b-exl2-6.5 --local-dir-use-symlinks False
Want to support my work? Visit my ko-fi page here: https://ko-fi.com/bartowski