39 lines
2.0 KiB
Markdown
39 lines
2.0 KiB
Markdown
# ollama-intel-gpu
|
|
|
|
Using Ollama for Intel based GPUs is not as straight forward as other natively Ollama supported platforms. As a workaround, this repo provides a quick sample showing the use of Ollama built with support for Intel ARC GPU based from the information provided by the references bellow. Run the recently released [Meta llama3](https://llama.meta.com/llama3) or [Microsoft phi3](https://news.microsoft.com/source/features/ai/the-phi-3-small-language-models-with-big-potential) models on your local Intel ARC GPU based PC using Linux or Windows WSL2.
|
|
|
|
## Screenshot
|
|
|
|
|
|
# Prerequisites
|
|
* Ubuntu 23.04 or newer (for Intel ARC GPU kernel driver support. Tested with Ubuntu 23.10), or Windows 11 with WSL2 (graphics driver [101.5445](https://www.intel.com/content/www/us/en/download/785597/intel-arc-iris-xe-graphics-windows.html) or newer)
|
|
* Installed Docker and Docker-compose tools (for Linux) or Docker Desktop (for Windows)
|
|
* Intel ARC series GPU (tested with Intel ARC A770 16GB)
|
|
|
|
# Usage
|
|
|
|
The following will build the Ollama with Intel ARC GPU support, and compose those with the public docker image based on OpenWEB UI from https://github.com/open-webui/open-webui
|
|
|
|
Linux:
|
|
```bash
|
|
$ git clone https://github.com/mattcurf/ollama-intel-gpu
|
|
$ cd ollama-intel-gpu
|
|
$ docker-compose up
|
|
```
|
|
|
|
Windows WSL2:
|
|
```bash
|
|
$ git clone https://github.com/mattcurf/ollama-intel-gpu
|
|
$ cd ollama-intel-gpu
|
|
$ docker-compose -f docker-compose-wsl2.yml up
|
|
```
|
|
|
|
Then launch your web browser to http://localhost:3000 to launch the web ui. Create a local OpenWeb UI credential, then click the settings icon in the top right of the screen, then select 'Models', then click 'Show', then download a model like 'llama3:8b-instruct-q8_0' for Intel ARC A770 16GB VRAM
|
|
|
|
# Known issues
|
|
* No effort has been made to prune the packages pulled into the Ollama docker image for Intel GPU
|
|
|
|
# References
|
|
* https://dgpu-docs.intel.com/driver/client/overview.html
|
|
* https://ipex-llm.readthedocs.io/en/latest/doc/LLM/Quickstart/ollama_quickstart.html
|