Update app.py

app.py

@@ -2,6 +2,8 @@ import gradio as gr
 import random
 import time
 from ctransformers import AutoModelForCausalLM
+from dl_hf_model import dl_hf_model
+
 
 params = {
         "max_new_tokens":512,
@@ -11,8 +13,9 @@ params = {
         "stream":True,
         "batch_size": 8}
 
-
-llm = AutoModelForCausalLM.from_pretrained("Aspik101/trurl-2-13b-GGML", model_type="llama")
+url = "https://huggingface.co/Aspik101/trurl-2-13b-GGML/blob/main/trurl-2-13b.ggmlv3.q8_0.bin" 
+model_loc, file_size = dl_hf_model(url)
+llm = AutoModelForCausalLM.from_pretrained(model_loc, model_type="llama")
 
 with gr.Blocks() as demo:
     chatbot = gr.Chatbot()
@@ -25,7 +28,8 @@ with gr.Blocks() as demo:
     def bot(history):
         print(history)
         #stream = llm(prompt =  f"Jesteś AI assystentem. Odpowiadaj po polski. <user>: {history}. <assistant>:", **params)
-        stream = llm(prompt =  f"{history}", **params)
+        #stream = llm(prompt =  f"{history}", **params)
+        stream = llm(prompt =  f"Jesteś AI assystentem. Odpowiadaj po polski. {history}.", **params)
         history[-1][1] = ""
         answer_save = ""
         for character in stream: