app.py

import gradio as gr
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("deepseek-ai/deepseek-llm-7b-base")
model = AutoModelForCausalLM.from_pretrained("deepseek-ai/deepseek-llm-7b-base", device_map="auto", torch_dtype="auto")

def chat_with_deepseek(message, history=[]):
    prompt = message
    inputs = tokenizer(prompt, return_tensors="pt").to(model.device)
    outputs = model.generate(**inputs, max_new_tokens=200)
    response = tokenizer.decode(outputs[0], skip_special_tokens=True)
    return response

iface = gr.Interface(
    fn=chat_with_deepseek,
    inputs="text",
    outputs="text",
    title="💬 DeepSeek Chat",
    description="Pose une question à DeepSeek AI !"
)

iface.launch()