Update app/services/llm_engine.py

app/services/llm_engine.py

@@ -119,7 +119,7 @@ class LLMEngine:
             # Kita panggil _execute_with_retry dan kirim 'messages' yang sudah lengkap tadi
             response_content = await self._execute_with_retry(
                 messages=messages,
-                model="llama-3.3-70b-versatile",
+                model="openai/gpt-oss-120b",
                 temperature=0.0,
                 response_format={"type": "json_object"}
             )
@@ -152,7 +152,7 @@ class LLMEngine:
         try:
             response_content = await self._execute_with_retry(
                 messages=[{"role": "user", "content": prompt}],
-                model="llama-3.3-70b-versatile",
+                model="openai/gpt-oss-120b",
                 temperature=0.5,
                 response_format={"type": "json_object"}
             )
@@ -185,7 +185,7 @@ class LLMEngine:
             # Perhatikan: Kita panggil _execute_with_retry, bukan self.client.create
             response_content = await self._execute_with_retry(
                 messages=[{"role": "user", "content": prompt}],
-                model="llama-3.3-70b-versatile",
+                model="openai/gpt-oss-120b",
                 response_format={"type": "json_object"}
             )
             return json.loads(response_content)
@@ -238,7 +238,7 @@ class LLMEngine:
             # PENTING: Panggil _execute_with_retry, JANGAN self.client
             response_content = await self._execute_with_retry(
                 messages=messages,
-                model="llama-3.1-8b-instant",
+                model="openai/gpt-oss-20b",
                 temperature=0.6 
             )
             return response_content
@@ -273,7 +273,7 @@ class LLMEngine:
         try:
             return await self._execute_with_retry(
                 messages=[{"role": "user", "content": prompt}],
-                model="llama-3.1-8b-instant",
+                model="openai/gpt-oss-20b",
                 temperature=0.7
             )
         except Exception as e:
@@ -310,7 +310,7 @@ class LLMEngine:
         try:
             return await self._execute_with_retry(
                 messages=[system_msg],
-                model="llama-3.1-8b-instant", 
+                model="openai/gpt-oss-20b", 
                 temperature=0.7
             )
         except Exception as e:
@@ -413,7 +413,7 @@ class LLMEngine:
         try:
             response_content = await self._execute_with_retry(
                 messages=[{"role": "system", "content": system_prompt}],
-                model="llama-3.3-70b-versatile",
+                model="openai/gpt-oss-120b",
                 temperature=0.3, # Rendah agar patuh data
                 response_format={"type": "json_object"}
             )