ok

Dockerfile

@@ -1,6 +1,5 @@
-FROM python:3.11-slim
+FROM python:3.10-slim
 
-# Set environment variables
 ENV PYTHONDONTWRITEBYTECODE=1 \
     PYTHONUNBUFFERED=1 \
     PORT=7860 \
@@ -8,34 +7,24 @@ ENV PYTHONDONTWRITEBYTECODE=1 \
 
 WORKDIR /app
 
-# Install system dependencies including build tools
+# Minimal system dependencies
 RUN apt-get update && apt-get install -y \
-    build-essential \
-    cmake \
     curl \
     && rm -rf /var/lib/apt/lists/*
 
-# Create non-root user (Required for HF Spaces)
+# Create non-root user (required for HF Spaces)
 RUN useradd -m -u 1000 user
 RUN mkdir -p /app/cache /app/models && chown -R user:user /app
 
-# Upgrade pip (as root before switching users)
-RUN pip install --no-cache-dir --upgrade pip
-
-# Switch to user
 USER user
 
-# Install llama-cpp-python with pre-built wheel
-RUN pip install --no-cache-dir llama-cpp-python \
-    --extra-index-url https://abetlen.github.io/llama-cpp-python/whl/cpu
-
-# Install the rest of the requirements
+# Install dependencies
+RUN pip install --no-cache-dir --upgrade pip
 COPY --chown=user:user requirements.txt .
 RUN pip install --no-cache-dir -r requirements.txt
 
-# Copy app code
+# Copy app
 COPY --chown=user:user main.py .
 
-# Run
 EXPOSE 7860
 CMD ["uvicorn", "main:app", "--host", "0.0.0.0", "--port", "7860"]

README.md

@@ -6,4 +6,6 @@ colorTo: pink
 sdk: docker
 app_file: main.py
 pinned: false
+python_version: "3.10"
+startup_duration_timeout: 1h
 ---

requirements.txt

@@ -1,3 +1,8 @@
+--prefer-binary
+--only-binary llama-cpp-python
+--extra-index-url https://abetlen.github.io/llama-cpp-python/whl/cpu
+
+llama-cpp-python>=0.2.90
 fastapi>=0.115.0
 uvicorn>=0.30.0
 pydantic>=2.8.0