File size: 2,329 Bytes
ff49c91
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
import os
import json
import gradio as gr
from llama_cpp import Llama

title = "Explore the Power of WestLake-7B: Advanced Conversational AI"
description = """
Welcome to the WestLake-7B Chat experience! Dive into the realm of advanced conversational AI with our cutting-edge model, WestLake-7B. 
This model, designed for nuanced and intelligent conversation, represents a significant leap in language processing capabilities. 
Explore its features and witness the future of AI-driven communication. 
Get your hands on the model and more details by clicking here: <a href="https://huggingface.co/senseable/Westlake-7B-gguf/">Westlake-7B on Hugging Face</a>.
"""

llm = Llama(model_path="model.gguf",
            n_ctx=32768,
            n_threads=2,
            chat_format="chatml")

def chat_stream_completion(message, history, system_prompt):
    messages_prompts = [{"role": "system", "content": system_prompt}]
    for human, assistant in history:
        messages_prompts.append({"role": "user", "content": human})
        messages_prompts.append({"role": "assistant", "content": assistant})
    messages_prompts.append({"role": "user", "content": message})

    response = llm.create_chat_completion(
        messages=messages_prompts,
        stream=True
    )
    message_repl = ""
    for chunk in response:
        if len(chunk['choices'][0]["delta"]) != 0 and "content" in chunk['choices'][0]["delta"]:
            message_repl = message_repl + chunk['choices'][0]["delta"]["content"]
        yield message_repl

gr.ChatInterface(
    fn=chat_stream_completion,
    title=title,
    description=description,
    additional_inputs=[gr.Textbox("Perform the task to the best of your ability.")],
    additional_inputs_accordion="📝 System prompt",
    examples=[
        ["Summarize the key themes of '1984' by George Orwell."],
        ["Convert 150 Fahrenheit to Celsius."],
        ["Generate a short story about a time-traveling detective."],
        ["Explain the concept of black holes in simple terms."],
        ["What are the health benefits of Mediterranean diet?"],
        ["Write Python code to sort a list of numbers in ascending order."],
        ["Translate 'Hello, how are you?' into French."],
        ["Create a motivational quote about perseverance."]
    ]

).queue().launch(server_name="0.0.0.0")