Intel / Step-3.5-Flash-int4-mixed-AutoRound

Model Details

This model is a mixed int4 model with group_size 128 and symmetric quantization of stepfun-ai/Step-3.5-Flash generated by intel/auto-round via RTN (no algorithm tuning). Please follow the license of the original model.

How To Use

INT4 Inference

start a vllm server:

vllm serve Intel/Step-3.5-Flash-int4-mixed-AutoRound \
  --served-model-name step3p5-flash-int4-mixed \
  --tensor-parallel-size 1 \
  --enable-expert-parallel \
  --disable-cascade-attn \
  --reasoning-parser step3p5 \
  --enable-auto-tool-choice \
  --tool-call-parser step3p5 \
  --hf-overrides '{"num_nextn_predict_layers": 1}' \
  --speculative_config '{"method": "step3p5_mtp", "num_speculative_tokens": 1}' \
  --trust-remote-code

Generate the Model

hf download stepfun-ai/Step-3.5-Flash --local-dir Step-3.5-Flash
auto_round ./Step-3.5-Flash --scheme W4A16 --iters 0 --disable_opt_rtn  --ignore_layers eh_proj,shared_head,layers.45 --layer_config "{mlp:{bits:8,data_type:int},self_attn:{bits:8,data_type:int},layers.46:{bits:8,data_type:int},layers.47:{bits:8,data_type:int}}"

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:

• Intel Neural Compressor
• AutoRound

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}
}

arxiv github