Merge pull request #141 from intel/update-branch

feat: update Dockerfile for new base image and Python environment setup; add CPU configuration instructions to README (#379)

Author: gooishin

usecases/ai/microservices/ollama/Dockerfile

@@ -1,31 +1,21 @@
 # Copyright (C) 2024 Intel Corporation
 # SPDX-License-Identifier: Apache-2.0
 
-FROM intel/oneapi-basekit:2024.2.1-0-devel-ubuntu22.04
+FROM intel/oneapi-basekit:2025.0.2-0-devel-ubuntu24.04
+ARG IPEX_LLM_VERSION=2.2.0b20250219
 
 WORKDIR /opt/intel/llm-app
-RUN rm -rf /etc/apt/sources.list.d/intel-graphics.list \
-    && apt update \
-    && apt install -y python3.11 \
-        python3.11-venv \
+RUN apt update \
+    && apt install -y python3-venv \
+        python3-pip \
         wget \
     && rm -rf /var/lib/apt/lists/* \
-    && python3.11 -m venv .venv
-
-RUN mkdir neo && cd neo \
-    && wget https://github.com/intel/intel-graphics-compiler/releases/download/v2.5.6/intel-igc-core-2_2.5.6+18417_amd64.deb \
-    && wget https://github.com/intel/intel-graphics-compiler/releases/download/v2.5.6/intel-igc-opencl-2_2.5.6+18417_amd64.deb \
-    && wget https://github.com/intel/compute-runtime/releases/download/24.52.32224.5/intel-level-zero-gpu-dbgsym_1.6.32224.5_amd64.ddeb \
-    && wget https://github.com/intel/compute-runtime/releases/download/24.52.32224.5/intel-level-zero-gpu_1.6.32224.5_amd64.deb \
-    && wget https://github.com/intel/compute-runtime/releases/download/24.52.32224.5/intel-opencl-icd-dbgsym_24.52.32224.5_amd64.ddeb \
-    && wget https://github.com/intel/compute-runtime/releases/download/24.52.32224.5/intel-opencl-icd_24.52.32224.5_amd64.deb \
-    && wget https://github.com/intel/compute-runtime/releases/download/24.52.32224.5/libigdgmm12_22.5.5_amd64.deb \
-    && dpkg -i *.deb
+    && python3 -m venv .venv
 
 ENV PATH="/opt/intel/llm-app/.venv/bin:$PATH"
 ENV LD_LIBRARY_PATH=".:$LD_LIBRARY_PATH"
 RUN python3 -m pip install --upgrade pip \
-    && python3 -m pip install --pre --upgrade 'ipex-llm[cpp]==2.2.0b20241216' \
+    && python3 -m pip install --pre --upgrade ipex-llm[cpp]==${IPEX_LLM_VERSION} \
     && python3 -m pip install --upgrade accelerate==0.33.0 \
     && init-ollama
 

usecases/ai/openwebui-ollama/README.md

@@ -117,6 +117,17 @@ Example:
 
     ![Alt Test](./assets/stt-example.gif)
 
+## FAQ
+### 1. Configure Ollama to use CPU instead of GPU
+* Linux: Export the environment variable `OLLAMA_NUM_GPU` before starting the services to offload to `CPU` device
+    ```bash
+    # Default: GPU
+    export OLLAMA_NUM_GPU=999
+
+    # Runs on CPU
+    export OLLAMA_NUM_GPU=0
+    ```
+
 ## Limitations
 ### 1. Automatic Speech Recognition Compatibility
 Automatic speech recognition functionality is not supported in Firefox. Please use Chrome for validated performance.

usecases/ai/openwebui-ollama/docker-compose.yml

@@ -88,9 +88,8 @@ services:
       - app-network
     environment:
       - OLLAMA_HOST=0.0.0.0:11434
-      - OLLAMA_NUM_GPU=999
+      - OLLAMA_NUM_GPU=${OLLAMA_NUM_GPU:-999}
       - ZES_ENABLE_SYSMAN=1
-      - SYCL_CACHE_PERSISTENT=1
       - SYCL_PI_LEVEL_ZERO_USE_IMMEDIATE_COMMANDLISTS=1
     devices:
       - /dev:/dev:rw