Update llm_handler.py

llm_handler.py

@@ -4,7 +4,7 @@ from llama_cpp_agent import MessagesFormatterType
 from llama_cpp_agent.providers import LlamaCppPythonProvider
 
 # Initialize the Llama model
-llama_model = Llama("Qwen2-1.5B-Instruct-Abliterated-iMat-GGUF/Qwen2-1.5B-Instruct-Abliterated-Q8_0.gguf", n_batch=1024, n_threads=24, n_gpu_layers=33, n_ctx=4098, verbose=False)
+llama_model = Llama("Qwen2-500M-Instruct-GGUF/Qwen2-500M-Instruct-f32.gguf", n_batch=1024, n_threads=24, n_gpu_layers=33, n_ctx=4098, verbose=False)
 
 # Create the provider
 provider = LlamaCppPythonProvider(llama_model)
@@ -19,7 +19,7 @@ agent = LlamaCppAgent(
 
 # Set provider settings
 settings = provider.get_provider_default_settings()
-settings.max_tokens = 2000
+settings.max_tokens = 4098
 settings.stream = True
 
 def send_to_llm(provider, msg_list):