diff --git a/docker-compose.coolify.yml b/docker-compose.coolify.yml
index 9c9afc7..3c685a2 100644
--- a/docker-compose.coolify.yml
+++ b/docker-compose.coolify.yml
@@ -164,8 +164,8 @@ services:
       test: ["CMD", "curl", "-f", "http://127.0.0.1:8095/health"]
       interval: 30s
       timeout: 10s
-      start_period: 120s
-      retries: 3
+      start_period: 300s
+      retries: 5
     restart: unless-stopped
     networks:
       - breakpilot-network
diff --git a/paddleocr-service/main.py b/paddleocr-service/main.py
index 7c1d581..b025f87 100644
--- a/paddleocr-service/main.py
+++ b/paddleocr-service/main.py
@@ -1,15 +1,20 @@
 """PaddleOCR Remote Service — PP-OCRv5 Latin auf x86_64."""
 
 import io
+import logging
 import os
 
 import numpy as np
 from fastapi import FastAPI, File, Header, HTTPException, UploadFile
 from PIL import Image
 
+logging.basicConfig(level=logging.INFO)
+logger = logging.getLogger(__name__)
+
 app = FastAPI(title="PaddleOCR Service")
 
 _engine = None
+_ready = False
 API_KEY = os.environ.get("PADDLEOCR_API_KEY", "")
 
 
@@ -18,6 +23,7 @@ def get_engine():
     if _engine is None:
         from paddleocr import PaddleOCR
 
+        logger.info("Loading PaddleOCR model (first time may download)...")
         _engine = PaddleOCR(
             lang="en",
             text_recognition_model_name="latin_PP-OCRv5_mobile_rec",
@@ -25,12 +31,27 @@ def get_engine():
             use_doc_unwarping=False,
             use_textline_orientation=False,
         )
+        logger.info("PaddleOCR model loaded successfully")
     return _engine
 
 
+@app.on_event("startup")
+def startup_load_model():
+    """Pre-load model at startup so health check passes."""
+    global _ready
+    try:
+        get_engine()
+        _ready = True
+        logger.info("PaddleOCR ready to serve requests")
+    except Exception as e:
+        logger.error(f"Failed to load PaddleOCR model: {e}")
+
+
 @app.get("/health")
 def health():
-    return {"status": "ok", "model": "PP-OCRv5-latin"}
+    if _ready:
+        return {"status": "ok", "model": "PP-OCRv5-latin"}
+    return {"status": "loading"}
 
 
 @app.post("/ocr")
@@ -41,6 +62,9 @@ async def ocr(
     if API_KEY and x_api_key != API_KEY:
         raise HTTPException(status_code=401, detail="Invalid API key")
 
+    if not _ready:
+        raise HTTPException(status_code=503, detail="Model still loading")
+
     img_bytes = await file.read()
     img = Image.open(io.BytesIO(img_bytes)).convert("RGB")
     img_np = np.array(img)