Update Dockerfile

Dockerfile

@@ -1,33 +1,22 @@
-# We use the full image, not slim, to ensure better compatibility
-FROM python:3.10-slim
+# ☢️ NUCLEAR OPTION: Use the Official Pre-Built Image
+# This image already has the AI engine (llama-cpp) compiled and installed.
+FROM ghcr.io/abetlen/llama-cpp-python:latest
 
-# 1. Setup User (Required by Spaces)
+# 1. Switch to Root to install our extra tools
+USER root
+
+# 2. Install only the missing pieces (FastAPI, Downloader)
+RUN pip install --no-cache-dir fastapi uvicorn huggingface_hub python-dotenv pydantic
+
+# 3. Setup the User (Strict Requirement for Hugging Face Spaces)
 RUN useradd -m -u 1000 user
 USER user
 ENV HOME=/home/user \
     PATH=/home/user/.local/bin:$PATH
-
 WORKDIR $HOME/app
 
-# 2. INSTALL THE FACTORY TOOLS (Mandatory)
-# We need these. Without them, the build fails.
-USER root
-RUN apt-get update && apt-get install -y \
-    build-essential \
-    cmake \
-    git \
-    && rm -rf /var/lib/apt/lists/*
-USER user
-
-# 3. INSTALL WITH FALLBACK
-# We try to download the binary. If that fails, we now have the tools (gcc) to build it.
-# We set CMAKE_ARGS to make the build faster if it happens.
-ENV CMAKE_ARGS="-DLLAMA_BLAS=OFF"
-
-COPY --chown=user requirements.txt requirements.txt
-RUN pip install --no-cache-dir --upgrade pip && \
-    pip install --no-cache-dir -r requirements.txt --extra-index-url https://abetlen.github.io/llama-cpp-python/whl/cpu
-
-# 4. LAUNCH
+# 4. Copy our "Librarian" Script
 COPY --chown=user . .
+
+# 5. Launch the Server
 CMD ["uvicorn", "main:app", "--host", "0.0.0.0", "--port", "7860"]