Spaces:
Sleeping
Sleeping
| import gradio as gr | |
| from huggingface_hub import InferenceClient | |
| import spaces | |
| """ | |
| 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("mkurman/llama-3.2-MEDIT-3B-o1") | |
| def respond( | |
| message, | |
| history: list[tuple[str, str]], | |
| # system_message, | |
| max_tokens, | |
| temperature, | |
| top_p, | |
| ): | |
| # messages = [{"role": "system", "content": system_message}] | |
| messages = [] | |
| 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 = "" | |
| counter = 0 | |
| for message in client.chat_completion( | |
| messages, | |
| max_tokens=max_tokens, | |
| stream=True, | |
| temperature=temperature, | |
| top_p=top_p, | |
| logprobs=True, | |
| stop=['</Output>'] | |
| ): | |
| text = message.choices[0].delta.content | |
| token = message.choices[0].logprobs.content[0].token | |
| if token in ['<Thought>', '<|python_tag|>']: | |
| text = '## Thinking:\n\n' + text | |
| if token == '<Output>': | |
| if counter > 0: | |
| text = '## Output:\n\n' + text | |
| else: | |
| text = '## Thinking:\n\n' + text | |
| if token == '</Output>': | |
| yield response | |
| break | |
| response += text | |
| counter += 1 | |
| yield response | |
| """ | |
| For information on how to customize the ChatInterface, peruse the gradio docs: https://www.gradio.app/docs/chatinterface | |
| """ | |
| demo = gr.ChatInterface( | |
| respond, | |
| title="LLama 3.2 MedIT 3B o1", | |
| description="Built with Llama. Please note that this model is not a source of knowledge and is not intended to provide 100% accurate answers. If the model provides answers that are generally considered inappropriate, please contact me.", | |
| additional_inputs=[ | |
| # gr.Textbox(value="You are a helpful, smart, kind, and efficient AI assistant. You always fulfill the user's requests to the best of your ability. You always think and reflect before providing final answers in a step-by-step manner.", label="System message"), | |
| gr.Slider(minimum=1, maximum=4096, value=2048, step=1, label="Max new tokens"), | |
| gr.Slider(minimum=0.0, maximum=2.0, value=0.1, 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() | |