server.py

from fastapi import FastAPI
from pydantic import BaseModel
import torch
from transformers import AutoTokenizer, AutoModelForCausalLM

app = FastAPI()

# Charge le modèle
model_name = "fatmata/psybot/psybot_model"  # Remplace par ton repo si différent
tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForCausalLM.from_pretrained(model_name)

class Message(BaseModel):
    text: str

@app.post("/chat")
def chat(message: Message):
    inputs = tokenizer(message.text, return_tensors="pt")
    output = model.generate(**inputs, max_length=100)
    response = tokenizer.decode(output[0], skip_special_tokens=True)
    return {"response": response}