from transformers import AutoModelForCausalLM, AutoTokenizer

model = AutoModelForCausalLM.from_pretrained(
    "nitrox/Jan-nano-128k-heretic",
    device_map="auto",
    torch_dtype="bfloat16",
)
tokenizer = AutoTokenizer.from_pretrained("nitrox/Jan-nano-128k-heretic")

Disclaimer

Trials: 500 (50 × 10 parallel GPUs)
Result: 0 refusals on eval set, KL divergence minimal

Refusal mechanisms have been removed. Use responsibly and in accordance with applicable laws.