nod-ai / SRT
Nod.ai π¦ version of π» . You probably want to start at https://github.com/nod-ai/shark for the product and the upstream IREE repository for mainline development. This repository houses branches and configuration that aren't ready for commit upstream.
β106Updated 2 months ago
Alternatives and similar repositories for SRT:
Users that are interested in SRT are comparing it to the libraries listed below
- Unified compiler/runtime for interfacing with PyTorch Dynamo.β99Updated 3 weeks ago
- β51Updated 7 months ago
- 3X speedup over Appleβs TensorFlow plugin by using Apache TVM on M1β135Updated 2 years ago
- Home for OctoML PyTorch Profilerβ108Updated last year
- Training neural networks in TensorFlow 2.0 with 5x less memoryβ130Updated 3 years ago
- MLIR-based partitioning systemβ73Updated this week
- benchmarking some transformer deploymentsβ26Updated 2 years ago
- An experimental CPU backend for Triton (https//github.com/openai/triton)β39Updated last week
- A thin, highly portable toolkit for efficiently compiling dense loop-based computation.β148Updated 2 years ago
- Stores documents and resources used by the OpenXLA developer communityβ117Updated 7 months ago
- extensible collectives library in tritonβ84Updated 6 months ago
- PyTorch interface for the IPUβ177Updated last year
- Repository for the QUIK project, enabling the use of 4bit kernels for generative inference - EMNLP 2024β177Updated 11 months ago
- Open source cross-platform compiler for compute-intensive loops used in AI algorithms, from Microsoft Researchβ110Updated last year
- β49Updated last year
- Backward compatible ML compute opset inspired by HLO/MHLOβ457Updated this week
- A Data-Centric Compiler for Machine Learningβ82Updated last year
- A tracing JIT compiler for PyTorchβ13Updated 3 years ago
- A lightweight, Pythonic, frontend for MLIRβ80Updated last year
- PyTorch RFCs (experimental)β131Updated 6 months ago
- A library of GPU kernels for sparse matrix operations.β260Updated 4 years ago
- torch::deploy (multipy for non-torch uses) is a system that lets you get around the GIL problem by running multiple Python interpreters iβ¦β178Updated 3 months ago
- The Foundation for All Legate Librariesβ207Updated this week
- β162Updated 9 months ago
- Customized matrix multiplication kernelsβ53Updated 3 years ago
- Fast sparse deep learning on CPUsβ52Updated 2 years ago
- A stand-alone implementation of several NumPy dtype extensions used in machine learning.β255Updated last week
- β73Updated 4 months ago
- A tensor-aware point-to-point communication primitive for machine learningβ256Updated 2 years ago
- Implementation of a Transformer, but completely in Tritonβ260Updated 2 years ago