Search Results for "cuda"
Sort By:
31 projects for "cuda" with 2 filters applied:
-
MongoDB Atlas runs apps anywhereMongoDB Atlas gives you the freedom to build and run modern applications anywhere—across AWS, Azure, and Google Cloud. With global availability in over 115 regions, Atlas lets you deploy close to your users, meet compliance needs, and scale with confidence across any geography.
-
Build Agents and Models on One PlatformGemini Enterprise Agent Platform is Google Cloud's comprehensive platform for developers to build, scale, govern, and optimize agents and models. Choose from Google's most advanced models and third-party models like Anthropic's Claude Model Family.
-
1
CUDA Agent
Large-Scale Agentic RL for High-Performance CUDA Kernel Generation
CUDA Agent is a research-driven agentic reinforcement learning system designed to automatically generate and optimize high-performance CUDA kernels for GPU workloads. The project addresses the long-standing challenge that efficient CUDA programming typically requires deep hardware expertise by training an autonomous coding agent capable of iterative improvement through execution feedback. -
2
how-to-optim-algorithm-in-cuda
How to optimize some algorithm in cuda
how-to-optim-algorithm-in-cuda is an open educational repository focused on teaching developers how to optimize algorithms for high-performance execution on GPUs using CUDA. The project combines technical notes, code examples, and practical experiments that demonstrate how common computational kernels can be optimized to improve speed and memory efficiency. 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. ... -
3
CUDA Containers for Edge AI & Robotics
Machine Learning Containers for NVIDIA Jetson and JetPack-L4T
CUDA Containers for Edge AI & Robotics is an open-source project that provides a modular container build system designed for running machine learning and AI workloads on NVIDIA Jetson devices. The repository contains container configurations that package the latest AI frameworks and dependencies optimized for Jetson hardware. These containers simplify the deployment of complex machine learning environments by bundling libraries such as CUDA, TensorRT, and deep learning frameworks into reproducible container images. ... -
4
GPU Puzzles
Solve puzzles. Learn CUDA
...Instead of presenting traditional lecture-style explanations, the project immerses learners directly in hands-on programming tasks that demonstrate how GPU computation works. The exercises are implemented using Python with the Numba CUDA interface, which allows Python code to compile into GPU kernels that run on CUDA-enabled hardware. By solving progressively more complex puzzles, learners gain a practical understanding of how parallel algorithms operate on graphics processing units. The project emphasizes experimentation and problem solving, encouraging learners to discover GPU programming techniques through trial and exploration. ... -
Compliant and Reliable File Transfers Backed by Top Security CertificationsStop relying on non-certified, legacy file transfer tools that creak under the weight of modern security demands. Get full audit trails, advanced access controls and more supported by an award-winning team of experts. Start your free 25-day trial today.
-
5
Model Zoo
Please do not feed the models
...The examples serve both as educational tools for learning Flux and as practical starting points for building new models. GPU acceleration is supported for most models through CUDA integration, enabling efficient training on compatible hardware. With community contributions encouraged, the Model Zoo acts as a hub for sharing and exploring diverse machine learning applications in Julia. -
6
VibeTensor
Our first fully AI generated deep learning system
VibeTensor is a groundbreaking open-source research system software stack for deep learning that was uniquely generated almost entirely by AI coding agents under guided human supervision, demonstrating a new frontier in AI-assisted software engineering. It implements a PyTorch-style eager tensor library with a modern C++20 core that supports both CPU and CUDA backends, giving it the ability to manage tensors, automatic differentiation (autograd), and complex computation flows similar to mainstream frameworks. What makes VibeTensor remarkable is that every major component, from core libraries and dispatch systems to CUDA runtime support, caching allocators, and language bindings, was created and validated by coding agents using automated builds and tests rather than manual line-by-line human coding. ... -
7
Koila
Prevent PyTorch's `CUDA error: out of memory` in just 1 line of code
...The system acts as a thin wrapper around PyTorch tensors and operations, meaning that it integrates easily into existing PyTorch code without requiring major changes to model implementations. It is particularly useful in environments where GPU resources are limited or where models frequently encounter CUDA memory errors. -
8
Lucebox
Fast LLM speculative inference server for consumer hardware
...The project includes a native C++ HTTP server with an OpenAI-compatible API, making it usable with tools that already speak the Chat Completions format. It supports CUDA and ROCm workflows, with Docker images for NVIDIA and AMD GPU setups. The repository also includes harnesses for testing compatibility with clients such as Claude Code, Codex, OpenCode, Hermes, Pi, OpenClaw, and Open WebUI. It is most useful for developers and AI enthusiasts who want to run optimized local models with lower latency, faster token generation, and hardware-aware inference behavior. -
9
DeepGEMM
Clean and efficient FP8 GEMM kernels with fine-grained scaling
DeepGEMM is a specialized CUDA library for efficient, high-performance general matrix multiplication (GEMM) operations, with particular focus on low-precision formats such as FP8 (and experimental support for BF16). The library is designed to work cleanly and simply, avoiding overly templated or heavily abstracted code, while still delivering performance that rivals expert-tuned libraries. -
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
Nano-vLLM
A lightweight vLLM implementation built from scratch
...The project recreates the core functionality of vLLM in a simplified architecture written in approximately a thousand lines of Python, making it easier for developers and researchers to understand how modern LLM inference systems work. Despite its compact design, nano-vllm incorporates advanced optimization techniques such as prefix caching, tensor parallelism, and CUDA graph execution to achieve high performance during model inference. The engine is intended primarily for educational use, experimentation, and lightweight deployments where a full production-grade inference stack may be unnecessary. Its API closely mirrors that of the original vLLM framework, allowing developers familiar with vLLM to adopt the tool with minimal changes. -
11
Open WebUI
User-friendly AI Interface
Open WebUI is an extensible, feature-rich, and user-friendly self-hosted AI platform designed to operate entirely offline. It supports various LLM runners like Ollama and OpenAI-compatible APIs, with a built-in inference engine for Retrieval Augmented Generation (RAG), making it a powerful AI deployment solution. Key features include effortless setup via Docker or Kubernetes, seamless integration with OpenAI-compatible APIs, granular permissions and user groups for enhanced security,... -
12
OuteTTS
Interface for OuteTTS models
...The project supports multiple backends including llama.cpp (Python bindings and server), Hugging Face Transformers, ExLlamaV2, VLLM and a JavaScript interface via Transformers.js, allowing it to run on CPUs, NVIDIA CUDA GPUs, AMD ROCm, Vulkan-capable GPUs, and Apple Metal. It also includes a notion of speaker profiles: you can create a speaker from a short audio sample, save it as JSON, and reuse it for consistent voice identity across generations and sessions. For best quality, the model is designed to work with a reference speaker clip and will inherit emotion, style, and accent from that reference. -
13
MLC LLM
Universal LLM Deployment Engine with ML Compilation
...The system supports deployment on environments including Linux, macOS, Windows, iOS, Android, and web browsers while utilizing different acceleration technologies such as CUDA, Vulkan, Metal, and WebGPU. It also provides OpenAI-compatible APIs that allow developers to integrate locally deployed models into existing AI applications without major code changes. -
14
DragonianVoice
C++ inference library for multiple SVC/TTS
...The project supports a wide range of model families: TTS models such as Tacotron2, VITS, EmotionalVITS, BERTVits2, GPT-SoVITS, SVC systems like SoVitsSvc (v2/v3/v4), RVC, DiffSvc, DiffusionSvc, FishDiffusion, ReflowSvc, and even singing systems like DiffSinger and related pitch/feature extractors like FCPE and RMVPE. It uses ONNX Runtime and other backends to accelerate inference, with notes on how different execution providers such as CUDA or DirectML affect operator support and numerical stability. Recent versions integrate with fish-speech via a dedicated fish-speech.cpp subproject using ggml. -
15
DeepSeek-V3.2-Exp
An experimental version of DeepSeek model
...MMLU, LiveCodeBench, AIME, Codeforces, etc.), V3.2-Exp shows performance very close to or in some cases matching that of V3.1-Terminus. The repository includes tools and kernels to support the new sparse architecture—for instance, CUDA kernels, logit indexers, and open-source modules like FlashMLA and DeepGEMM are invoked for performance. -
16
Instant Neural Graphics Primitives
Instant neural graphics primitives: lightning fast NeRF and more
Instant Neural Graphics Primitives, is an open-source research project developed by NVIDIA that enables extremely fast training and rendering of neural graphics representations. The system implements several neural graphics primitives including neural radiance fields, signed distance functions, neural images, and neural volumes. These representations are trained using a compact neural network combined with a multiresolution hash encoding that dramatically accelerates both training and... -
17
4M
4M: Massively Multimodal Masked Modeling
...The repository releases code and models for multiple variants (e.g., 4M-7 and 4M-21), emphasizing transfer to unseen tasks and modalities. Training/inference configs and issues discuss things like depth tokenizers, input masks for generation, and CUDA build questions, signaling active research iteration. The design leans into flexibility and steerability, so prompts and masks can shape behavior without bespoke heads per task. In short, 4M provides a unified recipe to pretrain large multimodal models that generalize broadly while remaining practical to fine-tune. -
18
SoniTranslate
Synchronized Translation for Videos
SoniTranslate is a video translation and dubbing system that produces synchronized target-language audio tracks for existing video content. It provides a web UI built with Gradio, allowing users to upload a video, choose source and target languages, and then run a pipeline that handles transcription, translation and re-synthesis of speech. Under the hood, it uses advanced speech and diarization models to separate speakers, align audio with timecodes and respect subtitle timing, which lets... -
19
CSM (Conversational Speech Model)
A Conversational Speech Generation Model
...The model has been fine-tuned for interactive voice demos and is hosted on platforms like Hugging Face for testing. CSM offers a flexible setup and is compatible with CUDA-enabled GPUs for efficient execution. -
20
Punica
Serving multiple LoRA finetuned LLM as one
...Punica introduces a serving architecture that allows multiple LoRA adapters to share the same base model during inference, significantly reducing memory consumption and computational overhead. The system includes specialized CUDA kernels that enable batched GPU operations across different LoRA models simultaneously. This design allows a single GPU cluster to host many task-specific models while maintaining high throughput and minimal latency. The architecture also includes scheduling mechanisms that coordinate requests from multiple tenants and distribute workloads efficiently across available resources. -
21
pipeless
A computer vision framework to create and deploy apps in minutes
Pipeless is an open-source computer vision framework to create and deploy applications without the complexity of building and maintaining multimedia pipelines. It ships everything you need to create and deploy efficient computer vision applications that work in real-time in just minutes. Pipeless is inspired by modern serverless technologies. It provides the development experience of serverless frameworks applied to computer vision. You provide some functions that are executed for new... -
22
Detectron
FAIR's research platform for object detection research
Detectron is an object detection and instance segmentation research framework that popularized many modern detection models in a single, reproducible codebase. Built on Caffe2 with custom CUDA/C++ operators, it provided reference implementations for models like Faster R-CNN, Mask R-CNN, RetinaNet, and Feature Pyramid Networks. The framework emphasized a clean configuration system, strong baselines, and a “model zoo” so researchers could compare results under consistent settings. It includes training and evaluation pipelines that handle multi-GPU setups, standard datasets, and common augmentations, which helped standardize experimental practice in detection research. ... -
23
SIG Rust
Rust language bindings for TensorFlow
SIG Rust provides idiomatic Rust bindings for TensorFlow, making it possible for developers to work with TensorFlow functionality from within the Rust programming language. Rather than replacing TensorFlow itself, it acts as an integration layer that connects Rust applications to the TensorFlow C API. The repository is designed for developers who want Rust’s performance, safety, and systems programming strengths while still accessing TensorFlow’s machine learning capabilities. It includes... -
24
LM Human Preferences
Code for the paper Fine-Tuning Language Models from Human Preferences
...The code is provided “as is” and explicitly says it may no longer run out-of-the-box due to dependencies or dataset migrations. It was tested on the smallest GPT-2 (124M parameters) under a specific environment (TensorFlow 1.x, specific CUDA / cuDNN combinations). It includes utilities for launching experiments, sampling from policies, and simple experiment orchestration. -
25
FasterTransformer
Transformer related optimization, including BERT, GPT
FasterTransformer is a high-performance inference library designed to accelerate transformer-based models such as BERT, GPT, and T5 on NVIDIA GPUs. It provides optimized implementations of transformer encoder and decoder layers using CUDA, cuBLAS, and custom kernels to maximize throughput and minimize latency. The library supports multiple deep learning frameworks, including TensorFlow, PyTorch, and Triton, allowing developers to integrate it into existing pipelines without major changes. It includes advanced optimization techniques such as mixed precision, tensor parallelism, and efficient memory management, enabling large models to run across multiple GPUs and nodes. ...
