Update app.py

app.py

@@ -7,7 +7,7 @@ from transformers import AutoModelForCausalLM, AutoTokenizer
 # -------------------------------------------------
 # Model setup (loaded once at startup)
 # -------------------------------------------------
-model_name = "CustomThinker-0-8B"
+model_name = "gr0010/CustomThinker-0-8B"
 
 # Load model and tokenizer globally
 print("Loading model and tokenizer...")
@@ -289,8 +289,11 @@ Think using bullet points and short sentences to simulate thoughts and emoticons
                 max_new_tokens=max_tokens
             )
 
-            # Update model history with CLEAN answer (no HTML formatting)
-            model_history.append({"role": "assistant", "content": answer})
+            # Update model history with thinking AND answer (CHANGED)
+            if thinking and thinking.strip():
+                model_history.append({"role": "assistant", "content": f"<think>{thinking}</think>\n{answer}"})
+            else:
+                model_history.append({"role": "assistant", "content": answer})
 
             # Format response for display (with HTML formatting)
             if thinking and thinking.strip():