Fix: update model names - auto-resolve

.env.example

@@ -33,7 +33,7 @@ ACCESS_TOKEN_EXPIRE_MINUTES=1440
 # --- GOOGLE GEMINI API (PRIMARY) ---
 # Get from: https://aistudio.google.com/app/apikey
 # You have Google Pro - this is your main LLM for response generation
-GOOGLE_API_KEY=AIzaSyXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX
+GOOGLE_API_KEY=your_google_api_key_here
 
 # Which Gemini model to use
 # Options: gemini-2.0-flash-lite, gemini-1.5-flash
@@ -47,10 +47,10 @@ GEMINI_TOKENS_PER_MINUTE=60000
 # --- GROQ API (SECONDARY) ---
 # Get from: https://console.groq.com/keys
 # Single key for specific fast inference tasks (llama models)
-GROQ_API_KEY=gsk_xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx
+GROQ_API_KEY=your_groq_api_key_here
 
 # Groq model (fast inference for policy evaluations)
-GROQ_MODEL=llama3-70b-8192
+GROQ_MODEL=llama-3.3-70b-versatile
 
 # Groq rate limits (Free tier)
 GROQ_REQUESTS_PER_MINUTE=30
@@ -60,7 +60,7 @@ GROQ_TOKENS_PER_MINUTE=30000
 # --- HUGGING FACE TOKEN (REQUIRED) ---
 # Get from: https://huggingface.co/settings/tokens
 # Required for: Model downloads (e5-base-v2, BERT), embeddings
-HF_TOKEN=hf_xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx
+HF_TOKEN=your_hf_token_here
 
 # ============================================================================
 # MODEL PATHS (Local storage)

app/config.py

@@ -47,8 +47,8 @@ class Settings:
     GROQ_API_KEY_3: str = os.getenv("GROQ_API_KEY_3", "")  # Fallback 2
     
     # Model names for Groq (using correct GroqCloud naming)
-    GROQ_CHAT_MODEL: str = os.getenv("GROQ_CHAT_MODEL", "llama3-8b-8192")  # For chat interface
-    GROQ_EVAL_MODEL: str = os.getenv("GROQ_EVAL_MODEL", "llama3-70b-8192")  # For evaluation
+    GROQ_CHAT_MODEL: str = os.getenv("GROQ_CHAT_MODEL", "llama-3.1-8b-instant")  # For chat interface
+    GROQ_EVAL_MODEL: str = os.getenv("GROQ_EVAL_MODEL", "llama-3.3-70b-versatile")  # For evaluation
     
     # ========================================================================
     # Commented as of now, can be re-enabled if rate limiting is needed
@@ -191,9 +191,9 @@ class Settings:
             str: Model name for the task
         """
         if task == "evaluation":
-            return self.GROQ_EVAL_MODEL  # llama3-70b-8192
+            return self.GROQ_EVAL_MODEL  # llama-3.3-70b-versatile
         else:
-            return self.GROQ_CHAT_MODEL  # llama3-8b-8192
+            return self.GROQ_CHAT_MODEL  # llama-3.1-8b-instant
 
 # ============================================================================
 # CREATE GLOBAL SETTINGS INSTANCE

app/core/llm_manager.py

@@ -35,8 +35,8 @@ class GroqManager:
     def __init__(self):
         """Initialize Groq manager with all available API keys"""
         self.api_keys = settings.get_groq_api_keys()
-        self.chat_model_name = settings.GROQ_CHAT_MODEL  # llama3-8b-8192
-        self.eval_model_name = settings.GROQ_EVAL_MODEL  # llama3-70b-8192
+        self.chat_model_name = settings.GROQ_CHAT_MODEL  # llama-3.1-8b-instant
+        self.eval_model_name = settings.GROQ_EVAL_MODEL  # llama-3.3-70b-versatile
         
         # Track current key index
         self.current_key_index = 0

backups/backup_config.py

@@ -52,7 +52,7 @@ class Settings:
     # GROQ API (Optional - for evaluation)
     # ========================================================================
     GROQ_API_KEY: str = os.getenv("GROQ_API_KEY", "")
-    GROQ_MODEL: str = os.getenv("GROQ_MODEL", "llama3-70b-8192")
+    GROQ_MODEL: str = os.getenv("GROQ_MODEL", "llama-3.3-70b-versatile")
     GROQ_REQUESTS_PER_MINUTE: int = int(os.getenv("GROQ_REQUESTS_PER_MINUTE", "30"))
     
     # ========================================================================
@@ -231,7 +231,7 @@ print("=" * 80)
 #     # GROQ API (Optional - for your llm_manager)
 #     # ========================================================================
 #     GROQ_API_KEY: str = os.getenv("GROQ_API_KEY", "")
-#     GROQ_MODEL: str = os.getenv("GROQ_MODEL", "llama3-70b-8192")
+#     GROQ_MODEL: str = os.getenv("GROQ_MODEL", "llama-3.3-70b-versatile")
     
 #     # ========================================================================
 #     # HUGGING FACE (Optional - for model downloads)
@@ -433,7 +433,7 @@ print("=" * 80)
 # #     # GROQ API (Optional - for your llm_manager)
 # #     # ========================================================================
 # #     GROQ_API_KEY: str = os.getenv("GROQ_API_KEY", "")
-# #     GROQ_MODEL: str = os.getenv("GROQ_MODEL", "llama3-70b-8192")
+# #     GROQ_MODEL: str = os.getenv("GROQ_MODEL", "llama-3.3-70b-versatile")
     
 # #     # ========================================================================
 # #     # HUGGING FACE (Optional - for model downloads)
@@ -563,7 +563,7 @@ print("=" * 80)
 #     # GROQ API (Optional - for your llm_manager)
 #     # ========================================================================
 #     GROQ_API_KEY: str = os.getenv("GROQ_API_KEY", "")
-#     GROQ_MODEL: str = os.getenv("GROQ_MODEL", "llama3-70b-8192")
+#     GROQ_MODEL: str = os.getenv("GROQ_MODEL", "llama-3.3-70b-versatile")
     
 #     # ========================================================================
 #     # HUGGING FACE (Optional - for model downloads)