wait for model loading on startup

run.sh

@@ -1,5 +1,6 @@
 #!/bin/bash
 set -e
+
 /llama.cpp/build/bin/llama-server \
   --hf-repo Qwen/Qwen2.5-7B-Instruct-GGUF \
   --hf-file qwen2.5-7b-instruct-q3_k_m.gguf \
@@ -12,10 +13,24 @@ set -e
 
 LLAMA_PID=$!
 
-sleep 8
+echo "Waiting for llama-server to become ready..."
+
+for i in $(seq 1 180); do
+  if curl -fsS http://127.0.0.1:8080/health >/dev/null; then
+    echo "llama-server is ready"
+    break
+  fi
+
+  if ! kill -0 "$LLAMA_PID" 2>/dev/null; then
+    echo "llama-server stopped during startup"
+    exit 1
+  fi
+
+  sleep 2
+done
 
-if ! kill -0 "$LLAMA_PID" 2>/dev/null; then
-  echo "llama-server failed to start"
+if ! curl -fsS http://127.0.0.1:8080/health >/dev/null; then
+  echo "llama-server did not become ready within 6 minutes"
   exit 1
 fi