app.py

from transformers import AutoModelForCausalLM, AutoTokenizer
import torch

model_name = "mistralai/Mistral-7B-v0.1"  # Örnek bir model

tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForCausalLM.from_pretrained(
    model_name,
    device_map="auto",
    load_in_8bit=True,  # 8-bit yükleme
    torch_dtype=torch.float16
)