libxsmm/tpp-mlir
TPP experimentation on MLIR for linear algebra
This project helps high-performance computing (HPC) and deep learning engineers optimize linear algebra operations. It takes in MLIR-based code and outputs highly optimized kernels that run efficiently on various CPU architectures. The ideal user is a system or performance engineer working on machine learning compilers or HPC libraries.
146 stars.
Use this if you are a system engineer or compiler developer looking to experiment with automatically selecting the best Tensor Processing Primitives for linear algebra within an MLIR framework to achieve higher performance.
Not ideal if you are an application developer simply looking to use an existing machine learning framework or library, as this is a low-level compiler infrastructure project.
Stars
146
Forks
38
Language
MLIR
License
—
Category
Last pushed
Mar 11, 2026
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/libxsmm/tpp-mlir"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Related frameworks
apache/tvm
Open Machine Learning Compiler Framework
uxlfoundation/oneDNN
oneAPI Deep Neural Network Library (oneDNN)
Tencent/ncnn
ncnn is a high-performance neural network inference framework optimized for the mobile platform
OpenMined/TenSEAL
A library for doing homomorphic encryption operations on tensors
iree-org/iree-turbine
IREE's PyTorch Frontend, based on Torch Dynamo.