Qwen3-14B-4bit-ASINQ

import torch
from transformers import AutoTokenizer
from sinq.patch_model import AutoSINQHFModel

model_name = "huawei-csl/Qwen3-14B-4bit-ASINQ"
tokenizer = AutoTokenizer.from_pretrained(model_name)
sinq_model = AutoSINQHFModel.from_quantized_safetensors(
    model_name,
    device="cuda:0",
    compute_dtype=torch.bfloat16
)

prompt = "Explain neural network quantization in one sentence."
inputs = tokenizer(prompt, return_tensors="pt").to("cuda:0")
with torch.inference_mode():
    out_ids = sinq_model.generate(**inputs, max_new_tokens=32, do_sample=False)
print(tokenizer.decode(out_ids[0], skip_special_tokens=True))

import torch
from transformers import AutoModelForCausalLM, AutoTokenizer
from sinq.patch_model import AutoSINQHFModel
from sinq.sinqlinear import BaseQuantizeConfig

# Load base model
base_model_name = "Qwen/Qwen3-14B"
model = AutoModelForCausalLM.from_pretrained(base_model_name, torch_dtype="float16")
tokenizer = AutoTokenizer.from_pretrained(base_model_name)

# Apply 4-bit SINQ quantization
quant_cfg = BaseQuantizeConfig(
    nbits=4,            # quantization bit-width
    group_size=64,     # group size
    tiling_mode="1D",   # tiling strategy
    method="asinq"       # quantization method ("asinq" for the calibrated version)
)

qmodel = AutoSINQHFModel.quantize_model(
    model,
    tokenizer=tokenizer,
    quant_config=quant_cfg,
    compute_dtype=torch.bfloat16,
    device="cuda:0"
)

@misc{muller2025sinq,
      title={SINQ: Sinkhorn-Normalized Quantization for Calibration-Free Low-Precision LLM Weights}, 
      author={Lorenz K. Muller and Philippe Bich and Jiawei Zhuang and Ahmet Celik and Luca Benfenati and Lukas Cavigelli},
      year={2025},
      eprint={2509.22944},
      archivePrefix={arXiv},
      primaryClass={cs.LG},
      url={http://arxiv.org/abs/2509.22944}
}