Software Development
ModelOpt Quantization-Aware Distillation
Run explicitly requested ModelOpt Quantization-Aware Distillation (QAD) through Megatron Bridge (recommended) or Hugging Face Transformers to recover a measured BF16-to-PTQ accuracy gap. Use only when the user explicitly asks for QAD, including its topology, data preparation, launch, resume, checkpoint export, or recovery decisions.
- Publisher / source
- NVIDIA/Model-Optimizer (found on GitHub top repositories)
- Upstream record
- NVIDIA/Model-Optimizer
- Version
- Version 1
- Published
- Publication review
- Approved for publication
- Content fingerprint
930855483582f61bf2ba85d40e696ea205974d2c9bc0d19171a791c0e8b0016a