neuralmagic/sparseml
A Python library for applying sparsification recipes (pruning, quantization) to neural networks with integrations for PyTorch, TensorFlow, and Keras.
Not currently ranked — collecting fresh signals.
star history
SparseML provides APIs and recipes for compressing neural networks to reduce their size and improve inference speed with minimal accuracy loss. It offers one-line integration with existing deep learning workflows and supports model optimization for computer vision, NLP, and other deep learning tasks. The library includes pre-built sparsification recipes and tools for creating custom optimization recipes targeting various model architectures.
Frequently asked
- What is neuralmagic/sparseml?
- A Python library for applying sparsification recipes (pruning, quantization) to neural networks with integrations for PyTorch, TensorFlow, and Keras.
- Is sparseml open source?
- Yes — neuralmagic/sparseml is open source, released under the Apache-2.0 license.
- What language is sparseml written in?
- neuralmagic/sparseml is primarily written in Python.
- How popular is sparseml?
- neuralmagic/sparseml has 2.1k stars on GitHub.
- Where can I find sparseml?
- neuralmagic/sparseml is on GitHub at https://github.com/neuralmagic/sparseml.