app.py

from transformers import AutoTokenizer, AutoModelForCausalLM
from peft import PeftModel

# Base model (pretrained Qwen)
base_model_name = "Qwen/Qwen2.5-7B-Instruct"

tokenizer = AutoTokenizer.from_pretrained(base_model_name)

# Load base model
base_model = AutoModelForCausalLM.from_pretrained(
    base_model_name,
    device_map="auto",
    trust_remote_code=True
)

# Load adapter weights on top
adapter_path = "gmacharla-team/qwen2.5b-finetuned"
model = PeftModel.from_pretrained(base_model, adapter_path)

# Now you can run inference
prompt = "Hello!"
inputs = tokenizer(prompt, return_tensors="pt")
outputs = model.generate(**inputs, max_new_tokens=100)
print(tokenizer.decode(outputs[0], skip_special_tokens=True))