Search Results for "network performance"
Sort By:
Showing 84 open source projects for "network performance"
View related business solutions-
Outplacement, Executive Coaching and Career Development | CareermindsBy helping to avoid the potential risks of RIFs or layoffs through our global outplacement services, companies can move forward with their goals while preserving their internal culture, employer brand, and bottom lines.
-
Dragonfly | An In-Memory Data Store without LimitsDragonfly is a drop-in Redis replacement that is designed for heavy data workloads running on modern cloud hardware. Migrate in less than a day and experience up to 25X the performance on half the infrastructure.
-
1
Velocity server
The modern, next-generation Minecraft server proxy
Velocity is a high-performance Minecraft proxy server developed by the PaperMC project that allows multiple Minecraft servers to be linked together into a single network. Acting as an intermediary between players and backend servers, Velocity manages player connections and routes them to different game servers within a network. This architecture allows large Minecraft communities to run multiple servers for different game modes while presenting them as a unified system to players. ... -
2
TensorRT
C++ library for high performance inference on NVIDIA GPUs
NVIDIA® TensorRT™ is an SDK for high-performance deep learning inference. It includes a deep learning inference optimizer and runtime that delivers low latency and high throughput for deep learning inference applications. TensorRT-based applications perform up to 40X faster than CPU-only platforms during inference. With TensorRT, you can optimize neural network models trained in all major frameworks, calibrate for lower precision with high accuracy, and deploy to hyperscale data centers, embedded, or automotive product platforms. ... -
3
ncnn
High-performance neural network inference framework for mobile
ncnn is a high-performance neural network inference computing framework designed specifically for mobile platforms. It brings artificial intelligence right at your fingertips with no third-party dependencies, and speeds faster than all other known open source frameworks for mobile phone cpu. ncnn allows developers to easily deploy deep learning algorithm models to the mobile platform and create intelligent APPs. -
4
Chrome DevTools MCP
Chrome DevTools for coding agents
...Beyond static inspection, it exposes operational tools like starting a performance trace that an agent can later analyze to propose optimizations. The server is intended to slot into MCP-capable assistants and IDEs, giving them reliable, typed tools and resource endpoints rather than ad-hoc automation. Documentation from the Chrome team explains how the server augments agents with real debugging capabilities. -
Junie, the AI coding agent by JetBrainsJunie is an AI-powered coding agent developed by JetBrains designed to enhance developer productivity by integrating directly into popular IDEs such as IntelliJ IDEA, PyCharm, and Android Studio. It supports developers by assisting with code completion, testing, and inspections, ensuring code quality and reducing debugging time.
-
5
imgclsmob Deep learning networks
Sandbox for training deep learning networks
...The project is frequently used by developers who want to study modern convolutional neural network designs and compare their performance across datasets. -
6
oneDNN
oneAPI Deep Neural Network Library (oneDNN)
This software was previously known as Intel(R) Math Kernel Library for Deep Neural Networks (Intel(R) MKL-DNN) and Deep Neural Network Library (DNNL). oneAPI Deep Neural Network Library (oneDNN) is an open-source cross-platform performance library of basic building blocks for deep learning applications. oneDNN is part of oneAPI. The library is optimized for Intel(R) Architecture Processors, Intel Processor Graphics and Xe Architecture graphics. oneDNN has experimental support for the following architectures: Arm* 64-bit Architecture (AArch64), NVIDIA* GPU, OpenPOWER* Power ISA (PPC64), IBMz* (s390x), and RISC-V. oneDNN is intended for deep learning applications and framework developers interested in improving application performance on Intel CPUs and GPUs. ... -
7
TEN Framework
TEN, a voice agent framework to create conversational AI.
TEN (Transformative Extensions Network) is a voice agent framework for creating conversational AI applications, focusing on high performance and modularity. -
8
face.evoLVe
High-Performance Face Recognition Library on PaddlePaddle & PyTorch
...The repository supports multiple neural network backbones such as ResNet, DenseNet, MobileNet, and ShuffleNet, enabling experimentation with different architectures depending on performance requirements. It also implements a wide range of loss functions commonly used in face recognition research, including ArcFace, CosFace, Triplet loss, and Softmax variants. To improve scalability, the library introduces distributed training techniques that allow large models to be trained efficiently across multiple GPUs. -
9
TurboQuant PyTorch
From-scratch PyTorch implementation of Google's TurboQuant
TurboQuant PyTorch is a specialized deep learning optimization framework designed to accelerate neural network inference and training through advanced quantization techniques within the PyTorch ecosystem. The project focuses on reducing the computational and memory footprint of models by converting floating-point representations into lower-precision formats while preserving performance. It provides tools for experimenting with different quantization strategies, enabling developers to balance accuracy and efficiency depending on their application. ... -
Managed Cybersecurity Platform Built for MSPsIn a world that lives and breathes all things digital, every business is at risk. Cybersecurity has become a major problem for small and growing businesses due to limited budgets, resources, time, and training. Hackers are leveraging these vulnerabilities, and most of the existing cybersecurity solutions on the market are too cumbersome, too complicated, and far too costly.
-
10
Intel Extension for PyTorch
A Python package for extending the official PyTorch
Intel® Extension for PyTorch* extends PyTorch* with up-to-date features optimizations for an extra performance boost on Intel hardware. Optimizations take advantage of Intel® Advanced Vector Extensions 512 (Intel® AVX-512) Vector Neural Network Instructions (VNNI) and Intel® Advanced Matrix Extensions (Intel® AMX) on Intel CPUs as well as Intel Xe Matrix Extensions (XMX) AI engines on Intel discrete GPUs. Moreover, Intel® Extension for PyTorch* provides easy GPU acceleration for Intel discrete GPUs through the PyTorch* xpu device. -
11
PowerInfer
High-speed Large Language Model Serving for Local Deployment
PowerInfer is a high-performance inference engine designed to run large language models efficiently on personal computers equipped with consumer-grade GPUs. The project focuses on improving the performance of local AI inference by optimizing how neural network computations are distributed between CPU and GPU resources. Its architecture exploits the observation that only a subset of neurons in large models are frequently activated, allowing the system to preload frequently used neurons into GPU memory while processing less common activations on the CPU. ... -
12
Edgee
AI gateway with token compression for Claude Code, Codex, and more
...It integrates AI capabilities into edge environments, making it possible to perform inference, personalization, and decision-making at the point of interaction. Edgee is optimized for performance and scalability, leveraging distributed execution to handle high volumes of requests efficiently. It also emphasizes developer experience, providing tools and abstractions that simplify deployment and orchestration across edge nodes. -
13
Gen.jl
A general-purpose probabilistic programming system
...But it also lets users migrate parts of their model or inference algorithm to specialized modeling languages for which it can generate especially fast code. Users can also hand-code parts of their models that demand better performance. Neural network inference is fast, but can be inaccurate on out-of-distribution data, and requires expensive training. -
14
COCOON
Confidential Compute Open Network, Decentralized AI Inference on TON
COCOON is a privacy-aware desktop client framework designed by the developers of Telegram to provide a modern, secure, and extensible environment for building messaging and communication applications. At its core, it combines native desktop performance with web-like flexibility, packing a renderer, UI components, and plugin architecture that allows developers to craft rich experiences similar to those found in native apps. Cocoon’s architecture prioritizes privacy and security, making it... -
15
NeuralNote
Audio Plugin for Audio to MIDI transcription using deep learning
...The software functions as an audio plugin that can be used inside digital audio workstations as well as a standalone application for music production and analysis. Its main purpose is to perform audio-to-MIDI transcription, allowing musicians to record a performance and automatically transform it into editable MIDI notes. NeuralNote supports polyphonic transcription, meaning it can detect multiple notes played simultaneously, making it useful for instruments such as piano or guitar. The system relies on neural network models to analyze audio signals and infer pitch, timing, and other musical attributes that can be represented as MIDI data. ... -
16
TurboQuant+
Implementation of TurboQuant (ICLR 2026)
...It is designed to be used in conjunction with modern machine learning workflows, particularly those involving large models that require optimization for deployment. TurboQuant Plus focuses on experimentation and performance tuning, allowing developers to test different configurations and evaluate trade-offs. Its architecture supports extensibility, enabling further development of quantization methods and integration with existing ML pipelines. -
17
BitNet
BitNet: Scaling 1-bit Transformers for Large Language Models
BitNet is a machine learning research implementation that explores extremely low-precision neural network architectures designed to dramatically reduce the computational cost of large language models. The project implements the BitNet architecture described in research on scaling transformer models using extremely low-bit quantization techniques. In this approach, neural network weights are quantized to approximately one bit per parameter, allowing models to operate with far lower memory usage than traditional 16-bit or 32-bit neural networks. ... -
18
Python Outlier Detection
A Python toolbox for scalable outlier detection
PyOD is a comprehensive and scalable Python toolkit for detecting outlying objects in multivariate data. This exciting yet challenging field is commonly referred as outlier detection or anomaly detection. PyOD includes more than 30 detection algorithms, from classical LOF (SIGMOD 2000) to the latest COPOD (ICDM 2020) and SUOD (MLSys 2021). Since 2017, PyOD [AZNL19] has been successfully used in numerous academic researches and commercial products [AZHC+21, AZNHL19]. PyOD has multiple neural... -
19
Attention Residuals (AttnRes)
Drop-in replacement for standard residual connections in Transformers
...Attention Residuals introduces a learnable softmax attention mechanism that allows each layer to selectively retrieve and weight useful representations from earlier layers, making depth dynamically adaptive rather than uniformly aggregated. This approach improves gradient stability, preserves meaningful signals throughout the network, and enhances performance in reasoning-heavy tasks such as coding, mathematics, and multi-step problem solving. -
20
UCCL
UCCL is an efficient communication library for GPUs
UCCL is a high-performance GPU communication library designed to support distributed machine learning workloads and large-scale AI systems. The library focuses on enabling efficient data transfer and collective communication between GPUs during training and inference processes. It supports a variety of communication patterns including collective operations such as all-reduce as well as peer-to-peer transfers that are commonly used in modern machine learning architectures. -
21
NNCF
Neural Network Compression Framework for enhanced OpenVINO
NNCF (Neural Network Compression Framework) is an optimization toolkit for deep learning models, designed to apply quantization, pruning, and other techniques to improve inference efficiency. -
22
daily.dev
Network for developers to learn, collaborate, and grow together
daily.dev is the open-source codebase behind the daily.dev developer news platform, which aggregates and personalizes technical content for programmers. The project focuses on delivering curated feeds of articles, tutorials, and community discussions tailored to developer interests. It combines content aggregation, ranking algorithms, and user personalization to create an engaging discovery experience. The platform is built with modern web technologies and emphasizes performance and... -
23
GPU Hot
Real-time NVIDIA GPU dashboard
GPU Hot is an open-source, lightweight monitoring dashboard designed to provide real-time visibility into NVIDIA GPU performance across single machines or entire clusters. The project offers a self-hosted web interface that streams hardware metrics directly from GPU servers, enabling developers, ML engineers, and system administrators to observe GPU utilization and system behavior in real time through a browser. The dashboard collects and displays a wide range of performance metrics... -
24
xLSTM
Neural Network architecture based on ideas of the original LSTM
xLSTM is an open-source machine learning architecture that reimagines the classic Long Short-Term Memory (LSTM) network for modern large-scale language modeling and sequence processing tasks. The project introduces a new recurrent neural network design that incorporates exponential gating mechanisms and enhanced memory structures to overcome limitations of traditional LSTM models. By introducing innovations such as matrix-based memory and improved normalization techniques, xLSTM improves the ability of recurrent networks to capture long-range dependencies in sequential data. ... -
25
ANE Training
Training neural networks on Apple Neural Engine via APIs
ANE Training is an experimental research project that demonstrates how to train neural networks directly on Apple’s Neural Engine by leveraging reverse-engineered private APIs that are normally inaccessible to developers. The repository implements a from-scratch transformer training pipeline capable of running both forward and backward passes on ANE hardware without relying on CoreML, Metal, or GPU acceleration. It explores the internal software stack of the Apple Neural Engine by...
