CePO

.dockerignore

@@ -0,0 +1,31 @@
+.git/
+.gitignore
+.github/
+__pycache__/
+*.pyc
+*.pyo
+*.pyd
+.Python
+*.so
+*.egg
+*.egg-info/
+dist/
+build/
+*.egg-info/
+.eggs/
+.tox/
+.coverage
+.coverage.*
+.cache
+nosetests.xml
+coverage.xml
+*.cover
+*.log
+.pytest_cache/
+.env
+.venv/
+env/
+venv/
+ENV/
+.idea/
+.vscode/

.github/workflows/publish.yml

@@ -1,5 +1,4 @@
-name: Upload Python Package to PyPI when a Release is Created
-
+name: Upload Python Package and Docker Image on Release
 on:
   release:
     types: [created]
@@ -25,6 +24,76 @@ jobs:
           pip install setuptools wheel
       - name: Build package
         run: |
-          python setup.py sdist bdist_wheel  # Could also be python -m build
+          python setup.py sdist bdist_wheel
       - name: Publish package distributions to PyPI
         uses: pypa/gh-action-pypi-publish@release/v1
+
+  docker-publish:
+    name: Publish Docker image
+    runs-on: ubuntu-latest
+    needs: pypi-publish
+    permissions:
+      contents: read
+      packages: write
+    steps:
+      - uses: actions/checkout@v4
+
+      - name: Set up QEMU
+        uses: docker/setup-qemu-action@v3
+
+      - name: Set up Docker Buildx
+        uses: docker/setup-buildx-action@v3
+
+      - name: Log in to GitHub Container Registry
+        uses: docker/login-action@v3
+        with:
+          registry: ghcr.io
+          username: ${{ github.actor }}
+          password: ${{ secrets.GITHUB_TOKEN }}
+
+      # Extract metadata for proxy_only image
+      - name: Extract metadata for proxy_only Docker
+        id: meta-proxy
+        uses: docker/metadata-action@v5
+        with:
+          images: ghcr.io/${{ github.repository }}
+          flavor: |
+            suffix=-slim
+          tags: |
+            type=semver,pattern={{version}}
+            type=semver,pattern={{major}}.{{minor}}
+            latest
+
+      # Build and push proxy image
+      - name: Build and push proxy_only Docker image
+        uses: docker/build-push-action@v5
+        with:
+          context: .
+          file: Dockerfile.proxy_only
+          push: true
+          platforms: linux/amd64,linux/arm64
+          tags: ${{ steps.meta-proxy.outputs.tags }}
+          labels: ${{ steps.meta-proxy.outputs.labels }}
+          cache-from: type=gha,scope=proxy
+          cache-to: type=gha,scope=proxy,mode=max
+
+      - name: Extract metadata for Docker
+        id: meta
+        uses: docker/metadata-action@v5
+        with:
+          images: ghcr.io/${{ github.repository }}
+          tags: |
+            type=semver,pattern={{version}}
+            type=semver,pattern={{major}}.{{minor}}
+            latest
+
+      - name: Build and push Docker image
+        uses: docker/build-push-action@v5
+        with:
+          context: .
+          push: true
+          platforms: linux/amd64,linux/arm64
+          tags: ${{ steps.meta.outputs.tags }}
+          labels: ${{ steps.meta.outputs.labels }}
+          cache-from: type=gha
+          cache-to: type=gha,mode=max

.gitignore

@@ -1,3 +1,5 @@
+# Apache license 2 - modified after the fork to add VS code 
+
 # Byte-compiled / optimized / DLL files
 __pycache__/
 *.py[cod]
@@ -164,3 +166,6 @@ cython_debug/
 # Ignore Mac DS_Store files
 .DS_Store
 **/.DS_Store
+
+# VS Code
+.vscode/

Dockerfile

@@ -1,12 +1,20 @@
 # Build stage
 FROM python:3.12-slim AS builder
 
+# Define build argument with default value
+ARG PORT=8000
+# Make it available as env variable at runtime
+ENV OPTILLM_PORT=$PORT
+
 # Set working directory
 WORKDIR /app
 
 # Install system dependencies
 RUN apt-get update && apt-get install -y --no-install-recommends \
-  gcc libc6-dev \
+  build-essential \
+  python3-dev \
+  gcc \
+  g++ \
   && rm -rf /var/lib/apt/lists/*
 
 # Copy only the requirements file first to leverage Docker cache
@@ -40,8 +48,8 @@ USER appuser
 # Set environment variables
 ENV PYTHONUNBUFFERED=1
 
-# Expose the port the app runs on
-EXPOSE 8000
+# Use the ARG in EXPOSE
+EXPOSE ${PORT}
 
 # Run the application
 ENTRYPOINT ["python", "optillm.py"]

Dockerfile.proxy_only

@@ -0,0 +1,55 @@
+# Build stage
+FROM python:3.12-slim AS builder
+
+# Define build argument with default value
+ARG PORT=8000
+# Make it available as env variable at runtime
+ENV OPTILLM_PORT=$PORT
+
+# Set working directory
+WORKDIR /app
+
+# Install system dependencies
+RUN apt-get update && apt-get install -y --no-install-recommends \
+  build-essential \
+  python3-dev \
+  gcc \
+  g++ \
+  && rm -rf /var/lib/apt/lists/*
+
+# Copy only the requirements file first to leverage Docker cache
+COPY requirements_proxy_only.txt .
+
+# Install Python dependencies
+RUN pip install --no-cache-dir -r requirements_proxy_only.txt
+
+# Final stage
+FROM python:3.12-slim
+
+# Install curl for the healthcheck
+RUN apt-get update && apt-get install -y --no-install-recommends \
+  curl && \
+  apt-get clean && rm -rf /var/lib/apt/lists/*
+
+# Set working directory
+WORKDIR /app
+
+# Copy installed dependencies from builder stage
+COPY --from=builder /usr/local/lib/python3.12/site-packages /usr/local/lib/python3.12/site-packages
+COPY --from=builder /usr/local/bin /usr/local/bin
+
+# Copy application code
+COPY . .
+
+# Create a non-root user and switch to it
+RUN useradd -m appuser
+USER appuser
+
+# Set environment variables
+ENV PYTHONUNBUFFERED=1
+
+# Use the ARG in EXPOSE
+EXPOSE ${PORT}
+
+# Run the application
+ENTRYPOINT ["python", "optillm.py"]

NOTICE.md

@@ -0,0 +1,24 @@
+## NOTICE
+
+This project is a fork of [https://github.com/codelion/optillm/tree/main](https://github.com/codelion/optillm/tree/main)
+
+The [original project](https://github.com/codelion/optillm/tree/main) is licensed under the Apache 2 License as detailed in [the original LICENSE](https://github.com/codelion/optillm/blob/main/LICENSE). The fork was created from a December 2024 fork of the original project.
+
+
+This project, i.e., [CePO](https://github.com/CerebrasResearch/cb_optillm/tree/cepo), is licensed under the Apache License, Version 2.0 as detailed in [this LICENSE](./LICENSE)
+
+
+Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the specific language governing permissions and limitations under the License.
+
+Initial change: 
+Adding Cerebras Planning and Optimization (CePO). CePO is a method to empower Llama with reasoning, via test-time compute. [Learn more in the blog](https://cerebras.ai/blog/cepo)
+
+At a high-level, in CePO, we make m attempts to generate n step-by-step plans, refine the plans, check inconsistencies between them, use the above feedback to generate the final plan and produce the answer. This process is then repeated N times in a classical best of n manner.
+
+Ongoing Apache-licensed contributions:
+* Added the implementation of CePO
+* Integrated CePO with optillm
+See updated files [here](https://github.com/codelion/optillm/compare/main...CerebrasResearch:cb_optillm:cepo)
+
+
+Last updated: 01/14/2025