modified

Dockerfile

@@ -1,12 +1,16 @@
 # Use the official vLLM image as the base image
 FROM vllm/vllm-openai:latest
 
-ENV TRANSFORMERS_CACHE="/tmp/huggingface_cache"
+# Set environment variables
+# ENV HUGGING_FACE_HUB_TOKEN="your_hf_token_here"
+ENV HF_HOME="/tmp/huggingface"
 ENV XDG_CACHE_HOME="/tmp/cache"
+ENV NUMBA_CACHE_DIR="/tmp/numba_cache"
+ENV OUTLINES_CACHE_DIR="/tmp/outlines_cache"
 
 # Create cache directories and set permissions
-RUN mkdir -p /tmp/huggingface_cache /tmp/cache && \
-    chmod 777 /tmp/huggingface_cache /tmp/cache
+RUN mkdir -p /tmp/huggingface /tmp/cache /tmp/numba_cache /tmp/outlines_cache && \
+    chmod -R 777 /tmp/huggingface /tmp/cache /tmp/numba_cache /tmp/outlines_cache
 
 # Set the working directory
 WORKDIR /app
@@ -21,7 +25,6 @@ EXPOSE 8000
 # Set the entrypoint
 ENTRYPOINT ["/app/entrypoint.sh"]
 
-
 # FROM python:3.9
 
 # # Create a non-root user

README.md

@@ -5,6 +5,8 @@ colorFrom: yellow
 colorTo: yellow
 sdk: docker
 pinned: false
+base_path: /v1
+app_port: 8000
 ---
 
 Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference