from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("Vynie/Serayuki-1B-v1.1-pre2-step-3k-1B")
model = AutoModelForCausalLM.from_pretrained("Vynie/Serayuki-1B-v1.1-pre2-step-3k-1B")

prompt = "Once upon a time, my friends and I wanted to"
inputs = tokenizer(prompt, return_tensors="pt").to("cuda")
outputs = model.generate(
    **inputs,
    max_length=256
)
print(tokenizer.decode(outputs[0], skip_special_tokens=True))

License

This project is licensed under the MIT License.