Update app/core/model_loader.py

app/core/model_loader.py

@@ -3,6 +3,7 @@ import json
 import logging
 import threading
 import time
+import random
 from datetime import datetime
 from dotenv import load_dotenv
 from groq import Groq
@@ -35,7 +36,11 @@ class LLMSingleton:
         else:
             self.client = Groq(api_key=self.api_key)
 
-        self.model_name = os.getenv("GROQ_MODEL", "llama3-8b-8192")
+        # Sanitize model string to strip hidden whitespace, newlines, or quotes injected by HF UI
+        raw_model = os.getenv("GROQ_MODEL", "llama-3.1-8b-instant")
+        clean_model = raw_model.strip().strip('"').strip("'") if raw_model else ""
+        self.model_name = clean_model if clean_model else "llama-3.1-8b-instant"
+
         self._stats_lock = threading.Lock()
         self._rpm_lock = threading.Lock()
 
@@ -125,8 +130,6 @@ class LLMSingleton:
         return True
 
     def generate(self, prompt: str, max_tokens: int = 2048) -> str:
-        import random
-
         self._check_daily_reset()
         self._check_rpm_window()