Update app.py

app.py

@@ -7,7 +7,7 @@ from llama_cpp import Llama
 from huggingface_hub import hf_hub_download #load from huggingfaces 
 
 
-llm = Llama(model_path= hf_hub_download(repo_id="TheBloke/Vigogne-2-7B-Chat-GGML", filename="vigogne-2-7b-chat.ggmlv3.q4_1.bin"), n_ctx=2048) #download model from hf/ n_ctx=2048 for high ccontext length
+llm = Llama(model_path= hf_hub_download(repo_id="TheBloke/orca_mini_3B-GGML", filename="orca-mini-3b.ggmlv3.q4_1.bin"), n_ctx=2048) #download model from hf/ n_ctx=2048 for high ccontext length
 
 history = []
 
@@ -17,7 +17,7 @@ def generate_text(input_text, history):
     if history == []:
         input_text_with_history = f"Q: {input_text} \n A:"
     else:
-        input_text_with_history = history[-1][1]+ "\n"
+        input_text_with_history = f"{history[-1][1]}"+ "\n"
         input_text_with_history += f"Q: {input_text} \n A:"
     print("new input", input_text_with_history)
     output = llm(input_text_with_history, max_tokens=1024, stop=["Q:", "\n"], stream=True)