Update Dockerfile

Dockerfile

@@ -3,13 +3,10 @@ FROM public.ecr.aws/q9t5s3a7/vllm-cpu-release-repo:latest
 
 # Avoid TRANSFORMERS_CACHE deprecation warning
 ENV HF_HOME=/opt/hf
+
 # Default CPU KV cache size (GiB) – tune for your RAM
 ENV VLLM_CPU_KVCACHE_SPACE=8
-# Default server host/port
-ENV HOST=0.0.0.0
-ENV PORT=8000
-# Model to serve – override at runtime with -e MODEL_ID=...
-ENV MODEL_ID=unsloth/Llama-3.2-3B-bnb-4bit
+
 # Extra args for vLLM
 ENV VLLM_ARGS="--dtype auto"
 
@@ -41,4 +38,10 @@ RUN vllm -v
 #      "--trust-remote-code", \
 #      "--device", "cpu"]
 
-CMD ["vllm", "serve", "unsloth/Llama-3.2-1B-bnb-4bit", "--host", "0.0.0.0", "--port", "7860", "--trust-remote-code", "--device", "cpu"]
+COPY start_server.sh /workspace
+
+WORKDIR /workspace
+
+ENTRYPOINT ["./start_server.sh"]
+
+# CMD ["vllm", "serve", "unsloth/Llama-3.2-1B-bnb-4bit", "--host", "0.0.0.0", "--port", "7860", "--trust-remote-code", "--device", "cpu"]