8000 -> 52415 (#915)

* 8000 -> 52415

* dont grab the api port for placement

---------

Co-authored-by: rltakashige <rl.takashige@gmail.com>

tmp/run_llm.py

@@ -7,7 +7,7 @@ import requests
 
 
 def stream_chat(host: str, query: str) -> None:
-    url = f"http://{host}:8000/v1/chat/completions"
+    url = f"http://{host}:52415/v1/chat/completions"
     headers = {"Content-Type": "application/json"}
     payload = {
         "model": "mlx-community/Llama-3.2-1B-Instruct-4bit",

tmp/run_llm.sh

@@ -10,7 +10,7 @@ HOST="$1"
 shift
 QUERY="$*"
 
-curl -sN -X POST "http://$HOST:8000/v1/chat/completions" \
+curl -sN -X POST "http://$HOST:52415/v1/chat/completions" \
   -H "Content-Type: application/json" \
   -d "{
         \"model\": \"mlx-community/Kimi-K2-Thinking\",