Statistics for topic cuda

A high-throughput and memory-efficient inference and serving engine for LLMs

The open-source serverless GPU container runtime.

CUDA on AMD GPUs

cuDF - GPU DataFrame Library

Official implementation of "Time Evidence Fusion Network: Multi-source View in Long-Term Time Series Forecasting" (https://arxiv.org/abs/2405.06419)

Official implementation of "Time Evidence Fusion Network: Multi-source View in Long-Term Time Series Forecasting" (https://arxiv.org/abs/2405.06419)

The open-source serverless GPU container runtime.

🚀 TensorRT-YOLO: Support YOLOv5, YOLOv6, YOLOv7, YOLOv8, YOLOv9, PP-YOLOE using TensorRT acceleration with EfficientNMS!

OneSweep, implemented in CUDA, D3D12, and Unity style compute shaders. Theoretically portable to all wave/warp/subgroup sizes.

Graphics Processing Units Molecular Dynamics

A high-throughput and memory-efficient inference and serving engine for LLMs

CUDA on AMD GPUs

The open-source serverless GPU container runtime.

World's fastest and most advanced password recovery utility

Samples for CUDA Developers which demonstrates features in CUDA Toolkit

Official implementation of "Time Evidence Fusion Network: Multi-source View in Long-Term Time Series Forecasting" (https://arxiv.org/abs/2405.06419)

The open-source serverless GPU container runtime.

🚀 TensorRT-YOLO: Support YOLOv5, YOLOv6, YOLOv7, YOLOv8, YOLOv9, PP-YOLOE using TensorRT acceleration with EfficientNMS!

OneSweep, implemented in CUDA, D3D12, and Unity style compute shaders. Theoretically portable to all wave/warp/subgroup sizes.

C++ open source 3D game engine

Official implementation of "Time Evidence Fusion Network: Multi-source View in Long-Term Time Series Forecasting" (https://arxiv.org/abs/2405.06419)

A high-throughput and memory-efficient inference and serving engine for LLMs

CUDA on AMD GPUs

A subset of PyTorch's neural network modules, written in Python using OpenAI's Triton.

World's fastest and most advanced password recovery utility

Instant neural graphics primitives: lightning fast NeRF and more

Official implementation of "Time Evidence Fusion Network: Multi-source View in Long-Term Time Series Forecasting" (https://arxiv.org/abs/2405.06419)

🚀 TensorRT-YOLO: Support YOLOv5, YOLOv6, YOLOv7, YOLOv8, YOLOv9, PP-YOLOE using TensorRT acceleration with EfficientNMS!

OneSweep, implemented in CUDA, D3D12, and Unity style compute shaders. Theoretically portable to all wave/warp/subgroup sizes.

A subset of PyTorch's neural network modules, written in Python using OpenAI's Triton.

The open-source serverless GPU container runtime.

Modular ZK(Zero Knowledge) backend accelerated by GPU

Drop-in, local AI alternative to the OpenAI stack. Multi-engine (llama.cpp, TensorRT-LLM). Powers 👋 Jan

Best inference performance optimization framework for HuggingFace Diffusers on NVIDIA GPUs.

3D Gaussian Splatting, reimagined: Unleashing unmatched speed with C++ and CUDA from the ground up!

Cross-platform, customizable multimedia/video processing framework. With strong GPU acceleration, heterogeneous design, multi-language support, easy to use, multi-framework compatible and high perfor...

A high-throughput and memory-efficient inference and serving engine for LLMs

Modular ZK(Zero Knowledge) backend accelerated by GPU

CUDA on AMD GPUs

Instant neural graphics primitives: lightning fast NeRF and more

World's fastest and most advanced password recovery utility

🎉CUDA 笔记 / 大模型手撕CUDA / C++笔记，更新随缘: flash_attn、sgemm、sgemv、warp reduce、block reduce、dot product、elementwise、softmax、layernorm、rmsnorm、hist etc.

CUDA Accelerated Robot Library

Open source, local, and self-hosted highly optimized language inference server supporting ASR/STT, TTS, and LLM across WebRTC, REST, and WS

CUDA C++ Core Libraries

ONNX Runtime Server: The ONNX Runtime Server is a server that provides TCP and HTTP/HTTPS REST APIs for ONNX inference.