Update lm.py

lm.py

@@ -382,32 +382,42 @@ if USE_LM_STUDIO:
     # CHAT
     # ---------------------------------------------------------
 
+    import re
+
     def _chat(prompt: str) -> str:
+     try:
+        response = _load_llm().create_chat_completion(
+            messages=[
+                {"role": "user", "content": prompt}
+            ],
+            temperature=0.3,
+            max_tokens=512,
+        )
 
-        try:
+        text = (
+            response["choices"][0]
+            .get("message", {})
+            .get("content", "")
+        )
 
-            model = _load_llm()
+        text = re.sub(
+            r"<think>.*?</think>",
+            "",
+            text,
+            flags=re.DOTALL
+        )
 
-            response = model.create_chat_completion(
-                messages=[
-                    {
-                        "role": "user",
-                        "content": prompt
-                    }
-                ],
-                temperature=0.3,
-                max_tokens=512,
-            )
+        text = text.replace("<start_of_turn>model", "")
+        text = text.replace("<end_of_turn>", "")
+        text = text.strip()
 
-            return (
-                response["choices"][0]
-                .get("message", {})
-                .get("content", "")
-            )
+        print("[lm] RESPONSE:", repr(text))
 
-        except Exception as e:
-            print("Chat error:", e)
-            return ""
+        return text
+
+     except Exception as e:
+        print("Chat error:", e)
+        return ""
 
     # ---------------------------------------------------------
     # EMBED