Update src/txagent/txagent.py

src/txagent/txagent.py

@@ -74,7 +74,7 @@ class TxAgent:
                 return f"The model {model_name} is already loaded."
             self.model_name = model_name
 
-        self.model = LLM(model=self.model_name, dtype="float16", max_model_len=2048, gpu_memory_utilization=0.8)
+        self.model = LLM(model=self.model_name, dtype="float16", max_model_len=32768, gpu_memory_utilization=0.8)
         self.chat_template = Template(self.model.get_tokenizer().chat_template)
         self.tokenizer = self.model.get_tokenizer()
         logger.info("Model %s loaded successfully", self.model_name)
@@ -386,8 +386,8 @@ class TxAgent:
                 temperature=temperature,
                 tools=picked_tools_prompt,
                 skip_special_tokens=False,
-                max_new_tokens=max_new_tokens,
-                max_token=max_token,
+                max_new_tokens=2048,
+                max_token=32768,
                 check_token_status=True)
             if last_outputs_str is None:
                 logger.warning("Token limit exceeded")
@@ -603,8 +603,8 @@ Summarize the function calls' responses in one sentence with all necessary infor
     def run_gradio_chat(self, message: str,
                         history: list,
                         temperature: float,
-                        max_new_tokens: int,
-                        max_token: int,
+                        max_new_tokens: 2048,
+                        max_token: 32768,
                         call_agent: bool,
                         conversation: gr.State,
                         max_round: int = 5,