FROM ghcr.io/ggml-org/llama.cpp:full WORKDIR /app RUN apt update && apt install -y python3 python3-pip python3-venv RUN python3 -m venv /opt/venv ENV PATH="/opt/venv/bin:$PATH" RUN pip install -U pip huggingface_hub # Model downloads RUN python3 -c 'from huggingface_hub import hf_hub_download; \ repo="HauhauCS/Gemma-4-E4B-Uncensored-HauhauCS-Aggressive"; \ hf_hub_download(repo_id=repo, filename="Gemma-4-E4B-Uncensored-HauhauCS-Aggressive-Q4_K_P.gguf", local_dir="/app"); \ hf_hub_download(repo_id=repo, filename="mmproj-Gemma-4-E4B-Uncensored-HauhauCS-Aggressive-f16.gguf", local_dir="/app")' # Blue-Purple CSS injection RUN find / -name "index.html" 2>/dev/null | xargs -I {} sed -i 's|||g' {} || true # Copy clean chat template into image COPY template.jinja /app/template.jinja CMD ["--server", \ "-m", "/app/Gemma-4-E4B-Uncensored-HauhauCS-Aggressive-Q4_K_P.gguf", \ "--mmproj", "/app/mmproj-Gemma-4-E4B-Uncensored-HauhauCS-Aggressive-f16.gguf", \ "--chat-template-file", "/app/template.jinja", \ "-r", "user", \ "-r", "", \ "--host", "0.0.0.0", \ "--port", "7860", \ "-t", "2", \ "--cache-type-k", "q8_0", \ "--cache-type-v", "iq4_nl", \ "-c", "8192", \ "-n", "4096"]