diff --git a/startup.sh b/startup.sh
index 009286e..3205780 100755
--- a/startup.sh
+++ b/startup.sh
@@ -2,9 +2,13 @@
 
 [ -f speech.env ] && . speech.env
 
-echo "First startup may download 2GB of speech models. Please wait."
+echo "First startup may download ~3GB of Qwen3-TTS model. Please wait."
 
-bash download_voices_tts-1.sh
-bash download_voices_tts-1-hd.sh $PRELOAD_MODEL
+# Pre-download Qwen3-TTS model (default engine)
+python -c "from qwen_tts import Qwen3TTSModel; Qwen3TTSModel.from_pretrained('Qwen/Qwen3-TTS-12Hz-1.7B-Base')" 2>/dev/null || echo "Qwen3-TTS will download on first request"
+
+# Optional: download legacy engines if enabled
+# bash download_voices_tts-1.sh
+# bash download_voices_tts-1-hd.sh $PRELOAD_MODEL
 
 python speech.py ${PRELOAD_MODEL:+--preload $PRELOAD_MODEL} $EXTRA_ARGS $@