File size: 3,069 Bytes
533133e
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
d5804bd
533133e
 
 
 
1a1ed94
533133e
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
f9caac8
533133e
 
 
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
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
from huggingface_hub import AsyncInferenceClient
import gradio as gr

client = AsyncInferenceClient("mistralai/Mixtral-8x7B-Instruct-v0.1")


def format_prompt(prompt: str, history: list[str], system_prompt: str) -> str:
    if not history:
        final_prompt = (
            f"[INST] {system_prompt if system_prompt else ''}:\n{prompt} [/INST]"
        )
    else:
        formatted_history = "".join(
            f"[INST] {user_prompt} [/INST]{bot_response}</s> "
            for user_prompt, bot_response in history
        )
        final_prompt = f"<s>{formatted_history}[INST] {prompt} [/INST]"
    return final_prompt


async def generate(
    prompt: str,
    history: list[str],
    system_prompt: str = "You're a helpful assistant.",
    temperature: float = 0.3,
    max_new_tokens: int = 4000,
    top_p: float = 0.95,
    repetition_penalty: float = 1.0,
):
    temperature = float(temperature)
    if temperature < 1e-2:
        temperature = 1e-2
    top_p = float(top_p)

    generate_kwargs = dict(
        temperature=temperature,
        max_new_tokens=max_new_tokens,
        top_p=top_p,
        repetition_penalty=repetition_penalty,
        do_sample=False,
        seed=42,
    )

    formatted_prompt = format_prompt(
        prompt=prompt, history=history, system_prompt=system_prompt
    )

    stream = await client.text_generation(
        formatted_prompt,
        **generate_kwargs,
        stream=True,
        details=True,
        return_full_text=True,
    )

    output = f""

    async for response in stream:
        output += response.token.text
        yield output


additional_inputs = [
    gr.Textbox(
        label="System Prompt (optional)",
        value="You're a helpful assistant.",
        info="This is experimental",
        placeholder="system prompt",
    ),
    gr.Slider(
        label="Temperature",
        value=0.9,
        minimum=0.0,
        maximum=1.0,
        step=0.05,
        interactive=True,
        info="Higher values produce more diverse outputs",
    ),
    gr.Slider(
        label="Max new tokens",
        value=256,
        minimum=0,
        maximum=1048,
        step=64,
        interactive=True,
        info="The maximum numbers of new tokens",
    ),
    gr.Slider(
        label="Top-p (nucleus sampling)",
        value=0.90,
        minimum=0.0,
        maximum=1,
        step=0.05,
        interactive=True,
        info="Higher values sample more low-probability tokens",
    ),
    gr.Slider(
        label="Repetition penalty",
        value=1.2,
        minimum=1.0,
        maximum=2.0,
        step=0.05,
        interactive=True,
        info="Penalize repeated tokens",
    ),
]

chatbot = gr.Chatbot(
    avatar_images=["./user.png", "./bot.png"],
    bubble_full_width=False,
    show_label=False,
    show_copy_button=True,
    likeable=True,
)

demo = gr.ChatInterface(
    fn=generate,
    additional_inputs=additional_inputs,
    chatbot=chatbot,
    title="🪷",
    description="Mixtral-8x7B-Instruct-v0.1",
    concurrency_limit=20,
)

demo.queue().launch()