Update app

app.py

@@ -6,29 +6,22 @@ import gradio as gr
 
 
 def generate_prompt(history):
-    prompt = start_message
+    prompt = "<s> "
     for chain in history[:-1]:
-        prompt += f"<|im_start|>user\n{chain[0]}<|im_end|>" +\
-                  f"<|im_start|>assistant\n{chain[1]}<|im_end|>"
-    prompt += f"<|im_start|>user\n{history[-1][0]}<|im_end|>" +\
-               "<|im_start|>assistant\n"
+        prompt += f"<human>: {chain[0]}\n<bot>: {chain[1]}{end_token}\n"
+    prompt += f"<human>: {history[-1][0]}\n<bot>: "
     return prompt
 
 def generate(history):
     prompt = generate_prompt(history)
 
-    streamer = llm(prompt, stop = stop_tokens, stream=True, threads=2)
+    streamer = llm(prompt, temperature=0, stream=True)
     return streamer
 
 
 
-llm = AutoModelForCausalLM.from_pretrained("model/ggml-model-q8_0.bin", model_type='mpt')
-stop_tokens = ["<|im_end|>", "<|endoftext|>"]
-
-start_message = """<|im_start|>system
-You are a helpful assistant chatbot.<|im_end|>
-"""
-
+llm = AutoModelForCausalLM.from_pretrained("model/model_q4_0.bin", model_type='llama')
+end_token = "</s>"
 
 
 with gr.Blocks() as demo:
@@ -53,4 +46,4 @@ with gr.Blocks() as demo:
     
 demo.queue()
 if __name__ == "__main__":
-    demo.launch(server_name="0.0.0.0", server_port=7860)
+    demo.launch()