Model Details
This model is an
mlx format 4.5b mixed model with group_size 128 and symmetric quantization of
Qwen/Qwen3.6-27B generated by
intel/auto-round. Please follow the license of the original model.
We currently support this format in AutoRound, but do not have the hardware to validate this large model.
As a result, we are unable to verify whether it runs correctly or achieves expected performance.
We would greatly appreciate your help in testing it, and welcome any contributions to our open-source project.
This model currently has some issues, as discussed in
ml-explore/mlx-lm#1214. For now, we recommend using AutoRound to generate uniform-bit models instead.
MLX-VLM inference
1from mlx_vlm import generate, load
2from mlx_vlm.prompt_utils import apply_chat_template
3from mlx_vlm.utils import load_config
4
5model_name_or_path= "Intel/Qwen3.6-27B-4.5b-mlx-AutoRound"
6
7model, processor = load(model_name_or_path)
8mlx_cfg = load_config(model_name_or_path)
9prompt_text = "Describe this image in one sentence."
10formatted = apply_chat_template(processor, mlx_cfg, prompt_text, num_images=1)
11# Use a public example image so the test does not need local assets.
12image_url = "https://qianwen-res.oss-accelerate.aliyuncs.com/Qwen3.5/demo/RealWorld/RealWorld-04.png"
13
14output = generate(model, processor, formatted, image=[image_url], max_tokens=2048).text
15print(output)
Generate the Model
1 AR_DISABLE_COPY_MTP_WEIGHTS=1 CUDA_VISIBLE_DEVICES=$device python3 -m auto_round \
2 --target_bits 4.5 \
3 --options "W4A16,W6A16,W8A16" \
4 --model_name $model_name \
5 --ignore_scale_zp_bits \
6 --format mlx \
7 --output_dir "./test_mlx_mixed" \
8 2>&1 | tee -a test_mlx.txt
Ethical Considerations and Limitations
The model can produce factually incorrect output, and should not be relied on to produce factually accurate information. Because of the limitations of the pretrained model and the finetuning datasets, it is possible that this model could generate lewd, biased or otherwise offensive outputs.
Therefore, before deploying any applications of the model, developers should perform safety testing.
Caveats and Recommendations
Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model.
Here are a couple of useful links to learn more about Intel's AI software:
Disclaimer
The license on this model does not constitute legal advice. We are not responsible for the actions of third parties who use this model. Please consult an attorney before using this model for commercial purposes.
Cite
@article{cheng2023optimize, title={Optimize weight rounding via signed gradient descent for the quantization of llms}, author={Cheng, Wenhua and Zhang, Weiwei and Shen, Haihao and Cai, Yiyang and He, Xin and Lv, Kaokao and Liu, Yi}, journal={arXiv preprint arXiv:2309.05516}, year={2023} }
