optimized files

Dockerfile

@@ -1,5 +1,11 @@
 FROM ollama/ollama
 
+# Pull model during build
+RUN ollama serve & \
+    sleep 10 && \
+    ollama pull embeddinggemma && \
+    pkill ollama
+
 COPY start.sh /start.sh
 
 RUN chmod +x /start.sh

start.sh

@@ -1,16 +1,7 @@
 #!/bin/sh
 
-# Tell Ollama to listen on port 7860
+# Listen on HF's expected port
 export OLLAMA_HOST=0.0.0.0:7860
 
-# Start server
-ollama serve &
-
-# Wait until server is ready
-sleep 10
-
-# Pull model
-ollama pull embeddinggemma
-
-# Keep container alive
-wait
+# Start Ollama in foreground
+exec ollama serve