edanor/umesimd
UME::SIMD A library for explicit simd vectorization.
Provides a unified abstraction layer across AVX, AVX2, AVX512, and Intel Xeon Phi (IMCI) instruction sets through homogeneous C++ interfaces, enabling portable SIMD code without backend-specific rewrites. Designed for explicit vectorization workflows where developers maintain fine-grained control over vector operations and register usage. Originally developed at CERN's ICE-DIP project for high-performance computing applications requiring cross-platform SIMD portability.
No commits in the last 6 months.
Stars
91
Forks
16
Language
C++
License
—
Category
Last pushed
Jan 19, 2018
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ai-coding/edanor/umesimd"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
symforce-org/symforce
Fast symbolic computation, code generation, and nonlinear optimization for robotics
inducer/loopy
A code generator for array-based code on CPUs and GPUs
alphaville/optimization-engine
Nonconvex embedded optimization: code generation for fast real-time optimization + ROS support
inducer/islpy
Python wrapper for isl, an integer set library
loco-3d/crocoddyl
Crocoddyl is an optimal control library for robot control under contact sequence. Its solver is...