Files
sglang/python/sglang/srt/layers/quantization/modelslim

Quantization ModelSlim module.

--quantization modelslim flag introduced. To load already quantized models, simply load the model weights. For models quantized with ModelSlim, there's no need to add --quantization modelslim argument when starting the engine. The quantization method will be automatically parsed from the downloaded quant_model_description.json config.

ModelSlim was developed in the format of compressed_tensors and includes support for various quantization schemes, such as:

  • W4A4 dynamic linear
  • W8A8 static linear
  • W8A8 dynamic linear
  • W4A8 dynamic MOE
  • W8A8 dynamic MOE

Also ModelSlim module include:

  • Automated config detection for modelslim format (without the need to specify --quantization modelslim flag)
  • Unit-tests for w4a4 modelslim, w8a8 modelslim