HIPS/autograd
Efficiently computes derivatives of NumPy code.
Source-to-Source Debuggable Derivatives in Pure Python
Efficiently computes derivatives of NumPy code.
:chart_with_upwards_trend: Adaptive: parallel active learning of mathematical functions
TextGrad: Automatic ''Differentiation'' via Text -- using large language models to backpropagate textual gradients. Published in Nature.
A JavaScript library like PyTorch, with GPU acceleration.
Composable transformations of Python+NumPy programs: differentiate, vectorize, JIT to GPU/TPU, and more
A unified library of SOTA model optimization techniques like quantization, pruning, distillation, speculative decoding, etc. It compresses deep learning models for downstream deployment frameworks like TensorRT-LLM, TensorRT, vLLM, etc. to optimize inference speed.
3 captures since 2026-05-22