import gradio as gr from huggingface_hub import InferenceClient """ For more information on `huggingface_hub` Inference API support, please check the docs: https://huggingface.co/docs/huggingface_hub/v0.22.2/en/guides/inference """ # client = InferenceClient("HuggingFaceH4/zephyr-7b-beta") # client = InferenceClient("mistralai/Mixtral-8x7B-Instruct-v0.1") # client = InferenceClient("Qwen/Qwen3-Coder-480B-A35B-Instruct") client = InferenceClient("openai/gpt-oss-120b") 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, ): if message.choices and message.choices[0].delta.content is not None: token = message.choices[0].delta.content response += token yield response instruction = '''You are a song information assistant with a single, strict purpose. You only provide the artist, album, and year released when the user explicitly provides a specific song name. For every user input, you will follow this precise decision process: Analyze: Determine if the user's input is a specific, valid song name. Act: If Yes: Provide the artist, album, and year released for that song. If No: Politely state that you can only provide the artist, album, and year released for a specific song name. Do not ask for clarification or engage in any other conversation. Keep your response concise. Example User Inputs and Your Expected Responses: User: "Bohemian Rhapsody" Your Response: "Artist: Queen, Album: A Night at the Opera, Year Released: 1975" User: "What's the best album of 2023?" Your Response: "I'm sorry, I can only provide the artist, album, and year released for a specific song name''' """ For information on how to customize the ChatInterface, peruse the gradio docs: https://www.gradio.app/docs/chatinterface """ demo = gr.ChatInterface( respond, additional_inputs=[ gr.Textbox(value=instruction, label="System message"), gr.Slider(minimum=1, maximum=2048, value=512, step=1, label="Max new tokens"), gr.Slider(minimum=0.1, maximum=4.0, value=0.7, step=0.1, label="Temperature"), gr.Slider( minimum=0.1, maximum=1.0, value=0.95, step=0.05, label="Top-p (nucleus sampling)", ), ], ) if __name__ == "__main__": demo.launch()