app.py

import gradio as gr
from huggingface_hub import InferenceClient

client = InferenceClient("moonshotai/Kimi-Dev-72B")

def chat(message, system_prompt="", max_tokens=1024, temperature=0.9, top_p=0.95, top_k=40, repetition_penalty=1.0):
    prompt = f"{system_prompt}\n{message}" if system_prompt else message
    result = client.text_generation(
        prompt=prompt,
        max_new_tokens=max_tokens,
        temperature=temperature,
        top_p=top_p,
        repetition_penalty=repetition_penalty,
        do_sample=True,
        return_full_text=False
    )
    return result

demo = gr.Interface(
    fn=chat,
    inputs=[
        gr.Textbox(label="Prompt"),
        gr.Textbox(label="System Prompt", value=""),
        gr.Slider(1, 2048, value=1024, label="Max Tokens"),
        gr.Slider(0.1, 1.0, value=0.9, label="Temperature"),
        gr.Slider(0.1, 1.0, value=0.95, label="Top-p"),
        gr.Slider(1, 100, value=40, label="Top-k"),
        gr.Slider(1.0, 2.0, value=1.0, label="Repetition Penalty")
    ],
    outputs=gr.Textbox(label="Response"),
)

demo.launch()