Search Results for "hardware" - Page 19
Sort By:
Showing 2171 open source projects for "hardware"
View related business solutions-
Ship Agents FasterGemini Enterprise Agent Platform lets you rapidly build, scale, govern and optimize production-ready agents grounded in your organization's data. The platform enables developers to build custom or pre-built agents for virtually any use case. New customers get $300 in free credits.
-
Cut Data Warehouse Costs by 54%BigQuery delivers 54% lower TCO with exabyte scale and flexible pricing. Free migration tools handle the SQL translation automatically.
-
1
PEFT
State-of-the-art Parameter-Efficient Fine-Tuning
Parameter-Efficient Fine-Tuning (PEFT) methods enable efficient adaptation of pre-trained language models (PLMs) to various downstream applications without fine-tuning all the model's parameters. Fine-tuning large-scale PLMs is often prohibitively costly. In this regard, PEFT methods only fine-tune a small number of (extra) model parameters, thereby greatly decreasing the computational and storage costs. Recent State-of-the-Art PEFT techniques achieve performance comparable to that of full... -
2
Mobly
E2E test framework for tests with complex environment requirements
Mobly is a Python-based test framework that specializes in supporting test cases that require multiple devices, complex environments, or custom hardware setups. P2P data transfer between two devices. Conference calls across three phones. Wearable device interacting with a phone. Internet-Of-Things devices interacting with each other. Testing RF characteristics of devices with special equipment. Testing LTE network by controlling phones, base stations, and eNBs. Mobly can support many different types of devices and equipment, and it's easy to plug your own device or custom equipment/service into Mobly. ... -
3
Graph Node
Graph Node indexes data from blockchains such as Ethereum
...This data can be transformed, organized, and shared across applications for anyone to query with just a few keystrokes. Before The Graph, teams had to develop and operate proprietary indexing servers. This required significant engineering and hardware resources and broke the important security properties required for decentralization. All data is stored and processed on open networks with verifiable integrity. The Graph makes querying this data fast, reliable, and secure. Entrepreneurs are creating next level apps to scale human coordination on the internet. It's a new frontier and we're just getting started. -
4
LMS plugins
Various plugins for Logitech Media Server
LMS is an open source multi-room audio system. It was created by SlimDevices and then acquired by Logitech. Although Logitech has ended development of hardware devices, the project is still very active with lots of plugin released regularly. New hardware players can be added using PI devices and DAC board to build high-quality multi-room. I've created some extra plugins, especially to include new standard hardware/players to the LMS ecosystem: UPnP/DLNA (Sonos), ChromeCast and AirPlay. -
Stop Cyber Threats with VM-Series Next-Gen Firewall on AzureGain integrated visibility across all traffic in a single pass. Deploy Palo Alto Networks VM-Series to determine application identity and content while automating security policy updates via rich APIs.
-
5
WebAssembly Micro Runtime
WebAssembly Micro Runtime (WAMR)
WebAssembly Micro Runtime (WAMR) is a lightweight, standalone runtime designed to execute WebAssembly modules efficiently across a wide range of environments, from embedded systems and IoT devices to cloud-native applications. Developed under the Bytecode Alliance, WAMR emphasizes a small memory footprint and high configurability, enabling it to operate effectively even in constrained hardware environments. The runtime supports multiple execution modes, including interpreter, Ahead-of-Time (AOT) compilation, and Just-in-Time (JIT) compilation, allowing developers to balance performance and portability depending on their needs. WAMR also provides near-native execution speed through its optimized compilation strategies, making it suitable for performance-critical applications. ... -
6
OpenFL
Open source library for creative expression on the web, desktop, etc.
OpenFL is a free and open-source, cross-platform software framework that empowers developers to create rich interactive applications and games using a single codebase that can run on web browsers, mobile devices, desktops, and even some consoles. It builds on the Haxe programming language and offers a familiar display list and event-driven API inspired by classic Adobe Flash and AIR, allowing developers to leverage well-known paradigms while targeting modern platforms. OpenFL supports 2D and... -
7
MegaTTS 3
Official PyTorch Implementation
MegaTTS3 is an open-source text-to-speech (TTS) and voice-cloning system from ByteDance that aims to deliver high-quality, expressive speech synthesis, including zero-shot voice cloning of previously unseen speakers. Its backbone is a lightweight diffusion-transformer (on the order of ~0.45 B parameters), which enables efficient inference while still producing high-fidelity audio. Given a reference audio sample (and corresponding latent representation), MegaTTS3 can generate speech in the... -
8
Style-Bert-VITS2
Style-Bert-VITS2: Bert-VITS2 with more controllable voice styles
...For those who only need synthesis, the project is published as a Python library (pip install style-bert-vits2) and can run on CPU without an NVIDIA GPU, though training still requires GPU hardware. -
9
CosyVoice
Multi-lingual large voice generation model, providing inference
CosyVoice is a multilingual large voice generation model that offers a full-stack solution for training, inference, and deployment of high-quality TTS systems. The model supports multiple languages, including Chinese, English, Japanese, Korean, and a range of Chinese dialects such as Cantonese, Sichuanese, Shanghainese, Tianjinese, and Wuhanese. It is designed for zero-shot voice cloning and cross-lingual or mix-lingual scenarios, so a single reference voice can be used to synthesize speech... -
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.
-
10
Xorbits Inference
Replace OpenAI GPT with another LLM in your app
Replace OpenAI GPT with another LLM in your app by changing a single line of code. Xinference gives you the freedom to use any LLM you need. With Xinference, you're empowered to run inference with any open-source language models, speech recognition models, and multimodal models, whether in the cloud, on-premises, or even on your laptop. Xorbits Inference(Xinference) is a powerful and versatile library designed to serve language, speech recognition, and multimodal models. With Xorbits... -
11
AIMET
AIMET is a library that provides advanced quantization and compression
...With this goal, QuIC presents the AI Model Efficiency Toolkit (AIMET) - a library that provides advanced quantization and compression techniques for trained neural network models. AIMET enables neural networks to run more efficiently on fixed-point AI hardware accelerators. Quantized inference is significantly faster than floating point inference. For example, models that we’ve run on the Qualcomm® Hexagon™ DSP rather than on the Qualcomm® Kryo™ CPU have resulted in a 5x to 15x speedup. Plus, an 8-bit model also has a 4x smaller memory footprint relative to a 32-bit model. However, often when quantizing a machine learning model (e.g., from 32-bit floating point to an 8-bit fixed point value), the model accuracy is sacrificed. -
12
RaspiBolt
Bitcoin & Lightning full node on a Raspberry Pi
Build your own “do-everything-yourself” Bitcoin full node that will make you a sovereign peer in the Bitcoin and Lightning network. With this guide, you can set up a Bitcoin and Lightning node from scratch, doing everything yourself. You will learn about Linux, Bitcoin, and Lightning. As a result, you’ll have your very own RaspiBolt node, built by you and no one else. Let your node validate your own Bitcoin transactions. No need to ask someone else to tell you what’s happening in the Bitcoin... -
13
YAO
Yao A low code engine to create web services and dashboard
Yao allows developers to create web services by processes. Yao is a low-code engine that creates a database model, writes API services, and describes dashboard interface just by JSON for web & hardware, no code, and 10x productivity. Yao is based on the flow-based programming idea, developed in the Go language, and supports multiple ways to expand the data stream processor. This makes Yao extremely versatile, which can replace programming languages in most scenarios, and is 10 times more efficient than traditional programming languages in terms of reusability and coding efficiency; application performance and resource ratio Better than PHP, JAVA and other languages. ... -
14
MiniMind-O
A 0.1B Omni model trained from scratch
...It extends the MiniMind family by exploring a model that can handle text, audio, and image inputs while producing text and streaming speech outputs. The project is designed to make multimodal AI training more accessible by keeping the model size small enough for ordinary personal hardware. It includes both mini and full training data paths, allowing learners to run a complete workflow quickly or reproduce the released model setup more closely. The implementation emphasizes native PyTorch code instead of relying on high-level third-party abstractions. minimind-o is most useful for developers and researchers who want to understand how multimodal and speech-capable AI systems are built from the ground up. -
15
QOI
The “Quite OK Image Format” for fast, lossless image compression
...QOI focuses on predictable O(n) performance and efficient pixel encoding strategies without relying on heavyweight compression algorithms like DEFLATE. The format supports RGB and RGBA images while remaining easy to port across programming languages and hardware environments. Its simplicity and speed have made it popular among game developers, embedded systems engineers, graphics programmers, and experimental software projects. Overall, qoi demonstrates how streamlined design can achieve practical high-performance image compression with dramatically reduced implementation complexity. -
16
ffmpeg-over-ip
Connect to remote ffmpeg servers
...The architecture includes authentication and command rewriting to adapt differences between client and server environments. It is designed to be flexible, supporting multiple clients connected to a single server instance. Overall, it provides an efficient way to extend hardware capabilities across networked systems. -
17
SIMD
C++ wrappers for SIMD intrinsics
...However, differences across CPU architectures and compilers make direct usage complex, which xsimd addresses by offering a unified API that maps efficiently to underlying hardware capabilities. The library supports a wide range of instruction sets, including SSE, AVX, NEON, and WebAssembly SIMD, ensuring portability across platforms. It provides vectorized implementations of common mathematical operations, allowing developers to operate on batches of values using familiar syntax. xsimd is widely adopted in performance-critical applications. -
18
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. Highway enables the same source code to run across different platforms and instruction sets, including architectures with scalable vector lengths, which are not known at compile time. This portability is achieved through dynamic or static dispatch mechanisms that select the best available instruction set at runtime or compile time. ... -
19
CUDA-QX
Accelerated libraries for quantum-classical computing built on CUDA-Q
...By abstracting low-level details and providing ready-to-use components, it accelerates experimentation and development in quantum computing research. The project is part of NVIDIA’s broader effort to enable scalable quantum-classical computing systems through hardware-agnostic programming models. -
20
Diffrax
Numerical differential equation solvers in JAX
...Because it is written to work closely with JAX, it supports just-in-time compilation, automatic differentiation, vectorization, and accelerator-backed execution on hardware such as GPUs and TPUs. This makes it especially appealing for researchers who need equation solvers that can be embedded inside trainable models or simulation-heavy learning systems. -
21
Gollama
Go manage your Ollama models
...One of its more distinctive capabilities is a VRAM estimation system that can calculate memory requirements, estimate context limits, and help users choose quantization settings that fit available hardware. -
22
llama.vim
Vim plugin for LLM-assisted code/text completion
...This approach allows developers to benefit from AI-assisted coding features while maintaining full control over their data and avoiding external API dependencies. The plugin focuses on simplicity and performance, providing fast completions and editing assistance even on consumer-grade hardware. By integrating AI functionality directly into Vim workflows, the tool enables developers to write and edit code more efficiently while staying within a familiar development interface. -
23
LLM-Finetuning
LLM Finetuning with peft
...The project focuses on parameter-efficient fine-tuning methods such as LoRA and QLoRA, which allow large models to be adapted to new tasks without requiring full retraining. Instead of requiring specialized hardware or complex training pipelines, many examples are designed to run in cloud notebook environments such as Google Colab. The repository includes step-by-step notebooks demonstrating how to fine-tune models such as LLaMA, Falcon, OPT, Vicuna, and GPT-NeoX. These tutorials show how developers can adapt pretrained models for tasks such as chatbots, classification, and instruction following. ... -
24
how-to-optim-algorithm-in-cuda
How to optimize some algorithm in cuda
...Instead of presenting only theoretical explanations, the repository includes hand-written CUDA implementations of fundamental operations such as reductions, element-wise computations, softmax, and attention mechanisms. These examples show how different optimization techniques influence performance on modern GPU hardware and allow readers to experiment with real implementations. The repository also contains extensive learning notes that summarize CUDA programming concepts, GPU architecture details, and performance engineering strategies. -
25
Megatron-LM
Ongoing research training transformer models at scale
...It supports advanced parallelism strategies including tensor, pipeline, data, expert, and context parallelism, enabling training across massive multi-GPU and multi-node clusters. The framework includes mixed-precision training options such as FP16, BF16, FP8, and FP4 to maximize performance and memory efficiency on modern hardware. Megatron-LM is widely used in research and industry for pretraining GPT-, BERT-, T5-, and multimodal-style models, with tooling for checkpoint conversion and interoperability with Hugging Face. Overall, it is a production-grade system for organizations pushing the limits of large-scale language model training.
