| | --- |
| | tags: |
| | - code |
| | - starcoder2 |
| | library_name: transformers |
| | pipeline_tag: text-generation |
| | license: bigcode-openrail-m |
| | quantized_by: bartowski |
| | --- |
| | |
| | ## Exllama v2 Quantizations of starcoder2-15b-instruct |
| |
|
| | Using <a href="https://github.com/turboderp/exllamav2/">turboderp's ExLlamaV2 v0.0.15 preview</a> 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/TechxGenus/starcoder2-15b-instruct |
| |
|
| | | Branch | Bits | lm_head bits | VRAM (4k) | VRAM (16k) | VRAM (32k) | Description | |
| | | ----- | ---- | ------- | ------ | ------ | ------ | ------------ | |
| | | [8_0](https://huggingface.co/bartowski/starcoder2-15b-instruct-exl2/tree/8_0) | 8.0 | 8.0 | 16.6 GB | 17.5 GB | 18.8 GB | Maximum quality that ExLlamaV2 can produce, near unquantized performance. | |
| | | [6_5](https://huggingface.co/bartowski/starcoder2-15b-instruct-exl2/tree/6_5) | 6.5 | 8.0 | 13.9 GB | 14.9 GB | 16.2 GB | Near unquantized performance at vastly reduced size, **recommended**. | |
| | | [5_0](https://huggingface.co/bartowski/starcoder2-15b-instruct-exl2/tree/5_0) | 5.0 | 6.0 | 11.2 GB | 12.2 GB | 13.5 GB | Slightly lower quality vs 6.5. | |
| | | [4_25](https://huggingface.co/bartowski/starcoder2-15b-instruct-exl2/tree/4_25) | 4.25 | 6.0 | 9.8 GB | 10.7 GB | 12.0 GB | GPTQ equivalent bits per weight. | |
| | | [3_5](https://huggingface.co/bartowski/starcoder2-15b-instruct-exl2/tree/3_5) | 3.5 | 6.0 | 8.4 GB | 9.3 GB | 10.6 GB | Lower quality, not recommended. | |
| | |
| | |
| | ## Download instructions |
| | |
| | With git: |
| | |
| | ```shell |
| | git clone --single-branch --branch 6_5 https://huggingface.co/bartowski/starcoder2-15b-instruct-exl2 |
| | ``` |
| | |
| | With huggingface hub (credit to TheBloke for instructions): |
| | |
| | ```shell |
| | pip3 install huggingface-hub |
| | ``` |
| | |
| | To download the `main` (only useful if you only care about measurement.json) branch to a folder called `starcoder2-15b-instruct-exl2`: |
| | |
| | ```shell |
| | mkdir starcoder2-15b-instruct-exl2 |
| | huggingface-cli download bartowski/starcoder2-15b-instruct-exl2 --local-dir starcoder2-15b-instruct-exl2 --local-dir-use-symlinks False |
| | ``` |
| | |
| | To download from a different branch, add the `--revision` parameter: |
| | |
| | Linux: |
| | |
| | ```shell |
| | mkdir starcoder2-15b-instruct-exl2-6_5 |
| | huggingface-cli download bartowski/starcoder2-15b-instruct-exl2 --revision 6_5 --local-dir starcoder2-15b-instruct-exl2-6_5 --local-dir-use-symlinks False |
| | ``` |
| | |
| | Windows (which apparently doesn't like _ in folders sometimes?): |
| |
|
| | ```shell |
| | mkdir starcoder2-15b-instruct-exl2-6.5 |
| | huggingface-cli download bartowski/starcoder2-15b-instruct-exl2 --revision 6_5 --local-dir starcoder2-15b-instruct-exl2-6.5 --local-dir-use-symlinks False |
| | ``` |
