from transformers import AutoModelForCausalLM, AutoTokenizer
from peft import PeftModel

# 1. Load the base model and tokenizer
base_model_id = "aisingapore/Llama-SEA-LION-v3-8B-IT"
tokenizer = AutoTokenizer.from_pretrained(base_model_id)
base_model = AutoModelForCausalLM.from_pretrained(base_model_id, device_map="auto", torch_dtype="auto")

# 2. Load your fine-tuned adapter
adapter_id = "jenniemoonsunset123/khmer-rag-model"
model = PeftModel.from_pretrained(base_model, adapter_id)

# 3. Test it!
prompt = "សួស្តី! តើអ្នកអាចជួយខ្ុំបានទេ?"
inputs = tokenizer(prompt, return_tensors="pt").to(model.device)
outputs = model.generate(**inputs, max_new_tokens=100)
print(tokenizer.decode(outputs[0], skip_special_tokens=True))