nick7nlp/RAZOR
RAZOR: Pruning Replaceable Experts in LLMs — training-free expert pruning for Mixture-of-Experts (MoE) models, with the MoECal calibration corpus
GitHub repository with 7 stars and 1 forks.
Language: Python
Topics: expert-pruning, llm, llm-compression, mixture-of-experts, model-compression, moe, moe-pruning, pruning, pytorch, training-free