Update Dockerfile

Dockerfile

@@ -4,47 +4,32 @@ ENV DEBIAN_FRONTEND=noninteractive
 ENV PYTHONUNBUFFERED=1
 ENV PIP_NO_CACHE_DIR=1
 
-# -------- Model config --------
-# Change these to your GGUF repo + file
 ENV MODEL_REPO_ID="rahul7star/Qwen3-4B-Thinking-2509-Genius-Coder-AI-Full"
 ENV MODEL_FILENAME="rahul7star_Qwen3-4B-Thinking-2509-Genius-Coder-AI-Full-Q5_K_M.gguf"
-ENV MODEL_NAME="hf.co/rahul7star/Qwen3-4B-Thinking-2509-Genius-Coder-AI-Full:Q5_K_M"
-# Cache/model dirs
 ENV HF_HOME="/data/.cache/huggingface"
 ENV HUGGINGFACE_HUB_CACHE="/data/.cache/huggingface"
 ENV MODEL_CACHE_DIR="/data/models"
-
-# App port for HF Spaces
-ENV PORT=7860
+ENV LLAMA_SERVER_URL="http://127.0.0.1:8000"
+ENV PORT="7860"
 
 RUN apt-get update && apt-get install -y --no-install-recommends \
     python3 \
     python3-pip \
     python3-venv \
-    build-essential \
-    git \
     curl \
     ca-certificates \
-    libopenblas-dev \
+    git \
     && rm -rf /var/lib/apt/lists/*
 
-WORKDIR /workspace
-
-COPY . /workspace/
+COPY --from=ghcr.io/ggml-org/llama.cpp:server /app/llama-server /usr/local/bin/llama-server
 
-RUN python3 -m pip install --upgrade pip setuptools wheel && \
-    pip3 install \
-      gradio==4.44.1 \
-      fastapi==0.112.2 \
-      starlette==0.38.6 \
-      pydantic==2.8.2 \
-      pydantic_core==2.20.1 \
-      "llama-cpp-python==0.2.90" \
-      "huggingface_hub==0.23.5"
+WORKDIR /workspace
 
-RUN chmod +x /workspace/entrypoint.sh && \
-    mkdir -p /data /workspace/data /data/models /data/.cache/huggingface
+COPY requirements.txt /workspace/requirements.txt
+RUN pip3 install --no-cache-dir -r requirements.txt
+RUN pip3 install --no-cache-dir requests huggingface_hub gradio
 
-EXPOSE 7860
+COPY . /workspace
+RUN chmod +x /workspace/entrypoint.sh
 
-CMD ["bash", "/workspace/entrypoint.sh"]
+CMD ["/workspace/entrypoint.sh"]