Top AI Repos — open-source AI, indexed and scored
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
Haste: a fast, simple, and open RNN library
| Date | Stars |
|---|---|
| 2026-07-24 | 334 |
| 2026-07-25 | 334 |
| 2026-07-28 | 334 |
| 2026-07-30 | 334 |
| 2026-08-06 | 334 |
Today
— stars today
This week
— stars this week
This month
— stars this month
Momentum
0.0
growth rate 0.00%/day
<div align="center"> <img src="https://lmnt.com/assets/haste-logo_social_media.png"> </div> -------------------------------------------------------------------------------- [](https://github.com/lmnt-com/haste/releases) [](https://colab.research.google.com/drive/1hzYhcyvbXYMAUwa3515BszSkhx1UUFSt) [](LICENSE) **We're hiring!** If you like what we're building here, [come join us at LMNT](https://explore.lmnt.com). Haste is a CUDA implementation of fused RNN layers with built-in [DropConnect](http://proceedings.mlr.press/v28/wan13.html) and [Zoneout](https://arxiv.org/abs/1606.01305) regularization. These layers are exposed through C++ and Python APIs for easy integration into your own projects or machine learning frameworks. Which RNN types are supported? - [GRU](https://en.wikipedia.org/wiki/Gated_recurrent_unit) - [IndRNN](http://arxiv.org/abs/1803.04831) - [LSTM](https://en.wikipedia.org/wiki/Long_short-term_memory) - [Layer Normalized GRU](https://arxiv.org/abs/1607.06450) - [Layer Normalized LSTM](https://arxiv.org/abs/1607.06450) What's included in this project? - a standalone C++ API (`libhaste`) - a TensorFlow Python API (`haste_tf`) - a PyTorch API (`haste_pytorch`) - examples for writing your own custom C++ inference / training code using `libhaste` - benchmarking programs to evaluate the performance of RNN implementations For questions or feedback about Haste, please open an issue on GitHub or send us an email at [[email protected]](mailto:[email protected]). ## Install Here's what you'll need to get started: - a [CUDA Compute Capability](https://developer.nvidia.com/cuda-gpus) 3.7+ GPU (required) - [CUDA Toolkit](https://developer.nvidia.com/cuda-toolkit) 10.0+ (required) - [TensorFlow GPU](https://www.tensorflow.org/install/gpu) 1.14+ or 2.0+ for TensorFlow integration (optional) - [PyTorch](https://pytorch.org) 1.3+ for PyTorch integration (optional) - [Eigen 3](http://eigen.tuxfamily.org/) to build the C++ examples (optional) - [cuDNN Developer Library](https://developer.nvidia.com/rdp/cudnn-archive) to build benchmarking programs (optional) Once you have the prerequisites, you can install with pip or by building the source code. ### Using pip ``` pip install haste_pytorch pip install haste_tf ``` ### Building from source ``` make # Build everything make haste # ;) Build C++ API make haste_tf # Build TensorFlow API make haste_pytorch # Build PyTorch API make examples make benchmarks ``` If you built the TensorFlow or PyTorch API, install it with `pip`: ``` pip install haste_tf-*.whl pip install haste_pytorch-*.whl ``` If the CUDA Toolkit that you're building against is not in `/usr/local/cuda`, you must specify the `$CUDA_HOME` environment variable before running make: ``` CUDA_HOME=/usr/local/cuda-10.2 make ``` ## Performance Our LSTM and GRU benchmarks indicate that Haste has the fastest publicly available implementation for nearly all problem sizes. The following charts show our LSTM results, but the GRU results are qualitatively similar. <table> <tr><td><img src="https://lmnt.com/assets/haste/benchmark/report_n=16_c=128.png"></td><td><img src="https://lmnt.com/assets/haste/benchmark/report_n=32_c=256.png"></td></tr> <tr></tr> <tr><td><img src="https://lmnt.com/assets/haste/benchmark/report_n=64_c=128.png"></td><td><img src="https://lmnt.com/assets/haste/benchmark/report_n=128_c=256.png"></td></tr> </table> Here is our complete LSTM benchmark result grid: <br> [`N=1 C=64`](https://lmnt.com/assets/haste/benchmark/report_n=1_c=64.png) [`N=1 C=128`](https://lmnt.com/assets/haste/benchmark/report_n=1_c=128.png) [`N=1 C=256`](https://lmnt.com/assets/haste/benchmark/report_n=1_c=256.png) [`N=1 C=512`](https://l
Excerpt of 9,549 characters
Read on GitHubWould you bet a product on this? Bounded 0–100 and slow moving.
matched fp:5114de53e4be560a, topic:deep-learning, topic:pytorch, topic:tensorflow