Tomoniai commited on
Commit
f50f605
1 Parent(s): 7702ac7

Create app.py

Browse files
Files changed (1) hide show
  1. app.py +30 -0
app.py ADDED
@@ -0,0 +1,30 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ import gradio as gr
2
+ from llama_cpp import Llama
3
+
4
+ llm = Llama(model_path="model.gguf", n_ctx=8000, n_threads=2, chat_format="chatml")
5
+
6
+ def generate(message, history,temperature=0.3,max_tokens=512):
7
+ system_prompt = "You are NeuralBeagle, a superintelligent AI assistant. "
8
+ formatted_prompt = [{"role": "system", "content": system_prompt}]
9
+ for user_prompt, bot_response in history:
10
+ formatted_prompt.append({"role": "user", "content": user_prompt})
11
+ formatted_prompt.append({"role": "assistant", "content": bot_response })
12
+ formatted_prompt.append({"role": "user", "content": message})
13
+ stream_response = llm.create_chat_completion(messages=formatted_prompt, temperature=temperature, max_tokens=max_tokens, stream=True)
14
+ response = ""
15
+ for chunk in stream_response:
16
+ if len(chunk['choices'][0]["delta"]) != 0 and "content" in chunk['choices'][0]["delta"]:
17
+ response += chunk['choices'][0]["delta"]["content"]
18
+ yield response
19
+
20
+ mychatbot = gr.Chatbot(
21
+ avatar_images=["user.png", "botnb.png"], bubble_full_width=False, show_label=False, show_copy_button=True, likeable=True,)
22
+
23
+ iface = gr.ChatInterface(fn=generate, chatbot=mychatbot, retry_btn=None, undo_btn=None)
24
+
25
+ with gr.Blocks() as demo:
26
+ gr.HTML("<center><h1>Tomoniai's Chat with Neural Beagle14-7b</h1></center>")
27
+ iface.render()
28
+
29
+ demo.queue().launch(show_api=False, server_name="0.0.0.0")
30
+