Runtime fallback: ensure /models/model.gguf exists, then exec llama-server

Dockerfile

@@ -4,3 +4,5 @@ ADD https://huggingface.co/bartowski/Qwen2.5-0.5B-Instruct-GGUF/resolve/main/Qwe
 
 EXPOSE 7860
 CMD ["-m","/models/model.gguf","-c","2048","-ngl","0","-t","4","--host","0.0.0.0","--port","7860"]
+ENTRYPOINT ["/bin/sh","-lc"]
+CMD ["test -f /models/model.gguf || (mkdir -p /models && curl -fL --retry 5 --retry-delay 2 -o /models/model.gguf https://huggingface.co/bartowski/Qwen2.5-0.5B-Instruct-GGUF/resolve/main/Qwen2.5-0.5B-Instruct-Q4_K_M.gguf); exec llama-server -m /models/model.gguf -c 2048 -ngl 0 -t 4 --host 0.0.0.0 --port 7860"]