File size: 2,952 Bytes
be82a8a
4971496
be82a8a
4971496
be82a8a
cdfe590
 
 
 
 
 
 
 
4971496
cdfe590
 
 
 
 
4971496
 
cdfe590
4971496
 
 
 
 
 
cdfe590
4971496
cdfe590
 
 
 
 
 
 
be82a8a
4971496
cdfe590
4971496
cdfe590
 
4971496
 
cdfe590
4971496
 
cdfe590
 
 
 
 
 
 
 
 
 
 
 
4971496
 
be82a8a
cdfe590
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
be82a8a
cdfe590
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
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
import gradio as gr
from huggingface_hub import InferenceClient

client = InferenceClient("Pinkstack/Superthoughts-lite-v1")

def respond(
    message,
    history: list[tuple[str, str]],
    system_message,
    max_tokens,
    temperature,
    top_p,
):
    messages = [{"role": "system", "content": system_message}]
    for val in history:
        if val[0]:
            messages.append({"role": "user", "content": val[0]})
        if val[1]:
            messages.append({"role": "assistant", "content": val[1]})
    messages.append({"role": "user", "content": message})
    response = ""
    for message in client.chat_completion(
        messages,
        max_tokens=max_tokens,
        stream=True,
        temperature=temperature,
        top_p=top_p,
    ):
        token = message.choices[0].delta.content
        response += token
        yield response

def format_response(response):
    # Replace <think>...</think> with a collapsible section
    response = response.replace("<think>", '<details><summary>Show thoughts</summary><div class="thoughts">')
    response = response.replace("</think>", "</div></details>")
    return response

css = """
.thoughts {
    border: 1px solid #ccc;
    padding: 10px;
    background-color: #f9f9f9;
    border-radius: 5px;
}
details summary {
    cursor: pointer;
    padding: 5px;
    background-color: #e0e0e0;
    border-radius: 5px;
    font-weight: bold;
}
details summary::-webkit-details-marker {
    display: none;
}
details summary:after {
    content: " ▶";
}
details[open] summary:after {
    content: " ▼";
}
"""

with gr.Blocks(css=css) as demo:
    gr.Markdown("## Chat with Superthoughts")
    gr.Markdown("**Warning:** The first output from the AI may take a few moments. After the first message, it should work quickly.")

    chatbot = gr.Chatbot()
    msg = gr.Textbox()
    system_message = gr.Textbox(value="You must always include <think> ... </think> <output> </output> tokens.", label="System message")
    max_tokens = gr.Slider(minimum=1, maximum=2048, value=512, step=1, label="Max new tokens")
    temperature = gr.Slider(minimum=0.1, maximum=4.0, value=0.7, step=0.1, label="Temperature")
    top_p = gr.Slider(minimum=0.1, maximum=1.0, value=0.95, step=0.05, label="Top-p (nucleus sampling)")

    def user(user_message, history):
        return "", history + [[user_message, None]]

    def bot(history, system_message, max_tokens, temperature, top_p):
        user_message, _ = history[-1]
        response = ""
        for partial_response in respond(user_message, history[:-1], system_message, max_tokens, temperature, top_p):
            response = partial_response
        formatted_response = format_response(response)
        history[-1][1] = formatted_response
        return history

    msg.submit(user, [msg, chatbot], [msg, chatbot], queue=False).then(
        bot, [chatbot, system_message, max_tokens, temperature, top_p], chatbot
    )

demo.launch()