Search Results for "simd"
Sort By:
Showing 80 open source projects for "simd"
View related business solutions-
Our Free Plans just got better! | Auth0You asked, we delivered! Auth0 is excited to expand our Free and Paid plans to include more options so you can focus on building, deploying, and scaling applications without having to worry about your security. Auth0 now, thank yourself later.
-
Go From AI Idea to AI App FastAccess Gemini 3 and 200+ models. Build chatbots, agents, or custom models with built-in monitoring and scaling.
-
1
SIMD
C++ wrappers for SIMD intrinsics
SIMD is a C++ library that provides portable abstractions over SIMD (Single Instruction, Multiple Data) instructions, enabling developers to write high-performance vectorized code without dealing directly with architecture-specific intrinsics. SIMD instructions allow a single operation to be applied to multiple data elements simultaneously, significantly accelerating numerical and data-parallel computations. -
2
Simd Library
C++ image processing and machine learning library with using of SIMD
...The algorithms are optimized with using of different SIMD CPU extensions. In particular, the library supports the following CPU extensions: SSE, AVX, AVX-512, and AMX for x86/x64, and NEON for ARM. The Simd Library has C API and also contains useful C++ classes and functions to facilitate access to C API. The library supports dynamic and static linking, 32-bit and 64-bit Windows and Linux, MSVS, G++ and Clang compilers, MSVS projects, and CMake build systems. -
3
Google Highway
Performance-portable, length-agnostic SIMD with runtime dispatch
Google Highway is a high-performance C++ library designed to provide portable SIMD (Single Instruction, Multiple Data) vectorization across multiple CPU architectures while maintaining predictable and efficient behavior. It abstracts low-level vector intrinsics into a consistent API that maps closely to hardware instructions, allowing developers to write high-performance code without relying heavily on compiler auto-vectorization. -
4
HLSL++
Math library using HLSL syntax with multiplatform SIMD support
...It provides vector, matrix, and math operations with a syntax identical or very similar to HLSL, allowing seamless transition between shader code and application code. The library is optimized for performance and supports SIMD instructions across multiple architectures, including SSE, AVX, AVX2, AVX512, and ARM NEON, ensuring high efficiency on modern hardware. It also extends beyond standard HLSL capabilities by introducing additional features such as quaternion support, advanced matrix operations, and extended vector types like float8. The library is particularly valuable for game developers who need consistency between CPU and GPU computations, reducing errors and improving maintainability. -
Train ML Models With SQL You Already KnowBuild and deploy ML models using familiar SQL. Automate data prep with built-in Gemini. Query 1 TB and store 10 GB free monthly.
-
5
LoopVectorization.jl
Macro(s) for vectorizing loops
LoopVectorization.jl is a Julia package for accelerating numerical loops by automatically applying SIMD (Single Instruction, Multiple Data) vectorization and other low-level optimizations. It analyzes loops and generates highly efficient code that leverages CPU vector instructions, making it ideal for performance-critical computing in fields such as scientific computing, signal processing, and machine learning. -
6
Datalevin
A simple, fast and versatile Datalog database
Datalevin is an open-source Datalog-based database written in Clojure that runs natively on top of LMDB. It supports full ACID transactions, schema-less EDN data storage, vector search with SIMD acceleration, and text-based querying. Usable as an embedded library or as a client/server database with RBAC access, it acts like SQLite or Datomic on a ledger of immutable datoms, plus modern features like vector and full-text search. -
7
ispc
Intel SPMD Program Compiler
...Under the SPMD model, the programmer writes a program that generally appears to be a regular serial program, though the execution model is actually that a number of program instances execute in parallel on the hardware. ispc compiles a C-based SPMD programming language to run on the SIMD units of CPUs and GPUs; it frequently provides a 3x or more speedup on architectures with 4-wide vector SSE units and 5x-6x on architectures with 8-wide AVX vector units, without any of the difficulty of writing intrinsics code. Parallelization across multiple cores is also supported by ispc, making it possible to write programs that achieve performance improvement that scales by both numbers of cores and vector unit size. ... -
8
QSV
Blazing-fast Data-Wrangling toolkit
qsv is a fast, command-line CSV data toolkit written in Rust that extends the capabilities of xsv. It’s designed to make working with CSV files at scale easy and efficient, offering over 40 powerful subcommands for tasks like querying, sampling, splitting, deduplicating, and more. qsv is ideal for data engineers, analysts, and developers who need high-performance CSV manipulation on the command line. -
9
node-rs
Node.js bindings Rust crates
When Node.js meets Rust. Make rust crates binding to Node.js use napi-rs. -
Application Monitoring That Won't Slow Your App DownFull APM with errors, performance, logs, and uptime monitoring. 99.999% uptime SLA on the platform itself.
-
10
Claude-Flow
The leading agent orchestration platform for Claude
...The platform supports both quick swarm tasks and persistent multi-agent sessions known as hives, facilitating distributed AI collaboration with persistent contextual memory. At its core, Claude-Flow integrates Dynamic Agent Architecture (DAA) for self-organizing agent management, neural pattern recognition accelerated by WebAssembly SIMD, and a SQLite-based memory system for context retention and knowledge persistence across tasks. It automates development workflows via pre- and post-operation hooks, providing seamless coordination, code formatting, validation, and performance optimization. -
11
Ultralight
Lightweight, high-performance HTML renderer for game developers
...Official API for C and C++, with bindings for more. Render web-content on the GPU via Direct3D, Metal, OpenGL, or your own engine for unmatched visual performance. Render web-content on the CPU via SIMD/parallel for incredibly easy integration with any environment (including server-side!). Ultralight is engineered for peak performance, ensuring minimal CPU and memory usage. Customize low-level platform functionality, integrate JavaScript directly with native code, dive deep into performance tuning, and more. Built for maximum portability, optimized for PCs, game consoles, TVs, and embedded systems. -
12
torchvision
Datasets, transforms and models specific to Computer Vision
The torchvision package consists of popular datasets, model architectures, and common image transformations for computer vision. We recommend Anaconda as Python package management system. Torchvision currently supports Pillow (default), Pillow-SIMD, which is a much faster drop-in replacement for Pillow with SIMD, if installed will be used as the default. Also, accimage, if installed can be activated by calling torchvision.set_image_backend('accimage'), libpng, which can be installed via conda conda install libpng or any of the package managers for debian-based and RHEL-based Linux distributions, and libjpeg, which can be installed via conda conda install jpeg or any of the package managers for debian-based and RHEL-based Linux distributions. ... -
13
Sonic JSON
A blazingly fast JSON serializing & deserializing library
A blazingly fast JSON serializing & deserializing library, accelerated by JIT (just-in-time compiling) and SIMD (single-instruction-multiple-data). -
14
Polars
Dataframes powered by a multithreaded, vectorized query engine
Polars is a high-performance, multi-language DataFrame library built in Rust using Apache Arrow. It delivers blazing-fast, vectorized, and parallel data manipulation with both eager and lazy execution, making it an excellent tool for data processing in Python, Rust, Node.js, R, and SQL contexts. -
15
wllama
WebAssembly binding for llama.cpp - Enabling on-browser LLM inference
...Built as a binding for the llama.cpp inference engine, the project allows developers to run LLM models locally without requiring a server backend or dedicated GPU hardware. The library leverages WebAssembly SIMD capabilities to achieve efficient execution within modern browsers while maintaining compatibility across platforms. By running models locally on the user’s device, wllama enables privacy-preserving AI applications that do not require sending data to remote servers. The framework provides both high-level APIs for common tasks such as text generation and embeddings, as well as low-level APIs that expose tokenization, sampling controls, and model state management. -
16
HighwayHash
Fast strong hash functions: SipHash/HighwayHash
HighwayHash is a fast, keyed hash function intended for scenarios where you need strong, DoS-resistant hashing without the full overhead of a general-purpose cryptographic hash. It’s designed to defeat hash-flooding attacks by mixing input with wide SIMD operations and a branch-free inner loop, so adversaries can’t cheaply craft many colliding keys. The implementation targets multiple CPU families with vectorized code paths while keeping a portable fallback, yielding high throughput across platforms. It exposes simple one-shot and streaming APIs, so you can hash short keys or long byte streams with the same function. ... -
17
ReverseDiff
Reverse Mode Automatic Differentiation for Julia
ReverseDiff is a fast and compile-able tape-based reverse mode automatic differentiation (AD) that implements methods to take gradients, Jacobians, Hessians, and higher-order derivatives of native Julia functions (or any callable object, really). While performance can vary depending on the functions you evaluate, the algorithms implemented by ReverseDiff generally outperform non-AD algorithms in both speed and accuracy. -
18
Numba
NumPy aware dynamic Python compiler using LLVM
Numba is an open source JIT compiler that translates a subset of Python and NumPy code into fast machine code. Numba translates Python functions to optimized machine code at runtime using the industry-standard LLVM compiler library. Numba-compiled numerical algorithms in Python can approach the speeds of C or FORTRAN. You don't need to replace the Python interpreter, run a separate compilation step, or even have a C/C++ compiler installed. Just apply one of the Numba decorators to your... -
19
Compute Library
The Compute Library is a set of computer vision and machine learning
The Compute Library is a set of computer vision and machine learning functions optimized for both Arm CPUs and GPUs using SIMD technologies. The library provides superior performance to other open-source alternatives and immediate support for new Arm® technologies e.g. SVE2. -
20
simdjson
Parsing gigabytes of JSON per second
JSON is everywhere on the Internet. Servers spend a *lot* of time parsing it. We need a fresh approach. The simdjson library uses commonly available SIMD instructions and microparallel algorithms to parse JSON 4x faster than RapidJSON and 25x faster than JSON for Modern C++. The simdjson library uses three-quarters less instructions than state-of-the-art parser RapidJSON. To our knowledge, simdjson is the first fully-validating JSON parser to run at gigabytes per second (GB/s) on commodity processors. ... -
21
Zerocopy
Zerocopy makes zero-cost memory manipulation effortless
Zerocopy is a Rust library designed to make zero-cost memory manipulation both safe and effortless. It allows developers to reinterpret or convert raw byte sequences into structured types—and vice versa—without writing unsafe code directly. The crate provides safe abstractions for transmuting data while preserving Rust’s strict safety guarantees, removing the need for manual memory manipulation. Zerocopy introduces a suite of conversion traits such as TryFromBytes, FromBytes, IntoBytes, and... -
22
StringZilla
10x faster string search, split, sort, and shuffle for long strings
...It matches the first few letters of words with hyper-scalar code to achieve memcpy speeds. The implementation fits into a single C 99 header file and uses different SIMD flavors and SWAR on older platforms. The Str is designed to replace long Python str strings and wrap our C-level API. On the other hand, the File memory-maps a file from persistent memory without loading its copy into RAM. The contents of that file would remain immutable, and the mapping can be shared by multiple Python processes simultaneously. ... -
23
Cactus
Low-latency AI inference engine optimized for mobile devices
Cactus is a low-latency, energy-efficient AI inference framework designed specifically for mobile devices and wearables, enabling advanced machine learning capabilities directly on-device. It provides a full-stack architecture composed of an inference engine, a computation graph system, and highly optimized hardware kernels tailored for ARM-based processors. Cactus emphasizes efficient memory usage through techniques such as zero-copy computation graphs and quantized model formats, allowing... -
24
QuestDB
An open source SQL database designed to process time series data
...These extensions make it simple to correlate data from multiple sources using relational and time series joins. QuestDB achieves high performance from a column-oriented storage model, massively-parallelized vector execution, SIMD instructions, and various low-latency techniques. The entire codebase was built from the ground up in Java and C++, with no dependencies, and is 100% free from garbage collection. We provide a live demo provisioned with the latest QuestDB release and sample datasets. -
25
XNNPACK
High-efficiency floating-point neural network inference operators
...Rather than serving as a standalone ML framework, XNNPACK provides high-performance computational primitives—such as convolutions, pooling, activation functions, and arithmetic operations—that are integrated into higher-level frameworks like TensorFlow Lite, PyTorch Mobile, ONNX Runtime, TensorFlow.js, and MediaPipe. The library is written in C/C++ and designed for maximum portability, efficiency, and performance, leveraging platform-specific instruction sets (e.g., NEON, AVX, SIMD) for optimized execution. It supports NHWC tensor layouts and allows flexible striding along the channel dimension to efficiently handle channel-split and concatenation operations without additional cost.
