Hugging Face's logo

Spaces:
Tonic
/
Phi-4
Runtime error

App Files Community
Phi-4
File size: 4,885 Bytes 
a94de3f
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
 
import gradio as gr
import transformers

title = """🙋🏻‍♂️Welcome to 🌟Tonic's 🤳🏻Phi-4 Demo"""

description = """
    This demo uses Microsoft's Phi-4 model for text generation.
    - System Prompt: Sets the context/role for the AI
    - User Prompt: Your specific question or request
    - Max Tokens: Maximum length of the generated response
    - Temperature: Controls randomness (higher = more creative, lower = more focused)
"""


join_us = """
## Join us:
🌟TeamTonic🌟 is always making cool demos! Join our active builder's 🛠️community 👻 
[![Join us on Discord](https://img.shields.io/discord/1109943800132010065?label=Discord&logo=discord&style=flat-square)](https://discord.gg/qdfnvSPcqP) 
On 🤗Huggingface: [MultiTransformer](https://huggingface.co/MultiTransformer) 
On 🌐Github: [Tonic-AI](https://github.com/tonic-ai) & contribute to🌟 [Dark Thoughts](https://github.com/MultiTonic/thinking-dataset)
🤗Big thanks to Yuvi Sharma and all the folks at huggingface for the community grant 🤗
"""


def generate_response(system_prompt, user_prompt, max_tokens, temperature):
    pipeline = transformers.pipeline(
        "text-generation",
        model="microsoft/phi-4",
        model_kwargs={"torch_dtype": "auto"},
        device_map="auto",
    )
    
    messages = [
        {"role": "system", "content": system_prompt},
        {"role": "user", "content": user_prompt},
    ]
    
    outputs = pipeline(
        messages, 
        max_new_tokens=max_tokens,
        temperature=temperature,
        do_sample=True
    )
    
    return outputs[0]["generated_text"]

# Example configurations
examples = [
    [
        "You are a medieval knight and must provide explanations to modern people.",
        "How should I explain the Internet?",
        128,
        0.7
    ],
    [
        "You are a wise wizard from ancient times.",
        "What would you call a smartphone?",
        256,
        0.8
    ],
    [
        "You are a time-traveling merchant from the year 1400.",
        "How would you describe modern cars?",
        200,
        0.6
    ],
    [
        "You are a medieval monk who specializes in manuscripts.",
        "What do you think about e-books?",
        150,
        0.7
    ],
    [
        "You are a castle guard from the Middle Ages.",
        "What do you think about modern security systems?",
        180,
        0.9
    ]
]

# Create the Gradio interface
with gr.Blocks() as demo:
    gr.Markdown(title)
    gr.Markdown(description)
    gr.Markdown(joinus)
    
    with gr.Row():
        with gr.Column():
            system_prompt = gr.Textbox(
                label="System Prompt",
                placeholder="Enter system prompt...",
                value="You are a medieval knight and must provide explanations to modern people."
            )
            user_prompt = gr.Textbox(
                label="User Prompt",
                placeholder="Enter your question...",
                value="How should I explain the Internet?"
            )
            
            with gr.Row():
                max_tokens = gr.Slider(
                    minimum=1,
                    maximum=512,
                    value=128,
                    step=1,
                    label="Maximum Tokens"
                )
                temperature = gr.Slider(
                    minimum=0.1,
                    maximum=1.0,
                    value=0.7,
                    step=0.1,
                    label="Temperature"
                )
            
            submit_btn = gr.Button("🚀 Generate Response")
        
        with gr.Column():
            output = gr.Textbox(
                label="Generated Response",
                lines=10
            )
    
    gr.Examples(
        examples=examples,
        inputs=[system_prompt, user_prompt, max_tokens, temperature],
        outputs=output,
        fn=generate_response,
        cache_examples=True,
        label="Example Prompts"
    )
    
    submit_btn.click(
        fn=generate_response,
        inputs=[system_prompt, user_prompt, max_tokens, temperature],
        outputs=output
    )
    
    gr.Markdown("""
    ### 📝 Parameters:
    - **System Prompt**: Sets the behavior/role of the AI (e.g., medieval knight, wizard, merchant)
    - **User Prompt**: Your question or input about modern concepts
    - **Maximum Tokens**: Controls the maximum length of the generated response
    - **Temperature**: Controls randomness (higher = more creative, lower = more focused)
    
    ### 💡 Tips:
    1. Try different historical personas in the system prompt
    2. Ask about modern technology from a historical perspective
    3. Adjust temperature for more varied or consistent responses
    4. Use the examples below for inspiration
    """)

# Launch the demo
if __name__ == "__main__":
    demo.launch()