from transformers import AutoModelForCausalLM, AutoTokenizer
import torch


model_name = "/Users/alpha/model/GPT-X2.5-135M"
tokenizer = AutoTokenizer.from_pretrained(model_name, trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained(
    model_name,
    trust_remote_code=True,
    dtype=torch.float32
)

prompt = "The main is"
inputs = tokenizer(prompt, return_tensors="pt").to(model.device)
with torch.inference_mode():
    output = model.generate(
        **inputs,
        max_new_tokens=120,
        temperature=0.8
    )

print(tokenizer.decode(output[0], skip_special_tokens=True))