Update app.py

app.py

@@ -908,34 +908,34 @@ REMEMBER: One tool per turn. No reasoning without tools. Exact answer format.
             #    max_new_tokens=2048, # Increased token limit for potentially longer reasoning/tool use
             #    temperature=0.01, # Keep temperature low for factual tasks     
             # Use tool_choice="any" to FORCE tool usage
-#        print("Initializing Groq LLM...")   
-#        try:
-#            self.llm_with_tools = ChatGroq(
-#                temperature=0,
-#                groq_api_key=GROQ_API_KEY,
-#                model_name="qwen/qwen3-32b",
-#                max_tokens=4096,
-#                timeout=60
-#                ).bind_tools(self.tools, tool_choice="auto")  
-#            print("✅ LLM initialized without FORCED tool usage.")
-#
-#        except Exception as e:
-#            print(f"❌ Error initializing HuggingFace: {e}")
-#            raise
-#        print("Initializing LLM Endpoint...")
-        llm = HuggingFaceEndpoint(
-            repo_id="Qwen/Qwen2.5-Coder-32B-Instruct", # Changed model
-            huggingfacehub_api_token=HUGGINGFACEHUB_API_TOKEN,
-            max_new_tokens=2048, # Increased token limit for potentially longer reasoning/tool use
-            temperature=0.01, # Keep temperature low for factual tasks
-        )
-        chat_llm = ChatHuggingFace(llm=llm)
-        print("✅ LLM Endpoint initialized.")
+        print("Initializing Groq LLM...")   
+        try:
+            self.llm_with_tools = ChatGroq(
+                temperature=0,
+                groq_api_key=GROQ_API_KEY,
+                model_name="qwen/qwen3-32b",
+                max_tokens=4096,
+                timeout=60
+                ).bind_tools(self.tools, tool_choice="auto")  
+            print("✅ LLM initialized without FORCED tool usage.")
 
-        # 6. Bind tools to the LLM
-        # Ensure the LLM knows how to format calls for the tools
-        self.llm_with_tools = chat_llm.bind_tools(self.tools)
-        print("✅ Tools bound to LLM.")
+        except Exception as e:
+            print(f"❌ Error initializing HuggingFace: {e}")
+            raise
+#        print("Initializing LLM Endpoint...")
+#        llm = HuggingFaceEndpoint(
+#            repo_id="Qwen/Qwen2.5-Coder-32B-Instruct", # Changed model
+#            huggingfacehub_api_token=HUGGINGFACEHUB_API_TOKEN,
+#            max_new_tokens=2048, # Increased token limit for potentially longer reasoning/tool use
+#            temperature=0.01, # Keep temperature low for factual tasks
+#        )
+#        chat_llm = ChatHuggingFace(llm=llm)
+#        print("✅ LLM Endpoint initialized.")
+#
+#        # 6. Bind tools to the LLM
+#        # Ensure the LLM knows how to format calls for the tools
+#        self.llm_with_tools = chat_llm.bind_tools(self.tools)
+#        print("✅ Tools bound to LLM.")
 #        print("Initializing Google Gemini LLM...")
 #        try:
 #            self.llm_with_tools = ChatGoogleGenerativeAI(