Kashipara.com is a community of ONE million programmers and students, Just like you, Helping each other.Join them. It only takes a minute: Sign Up
Job Resume TemplateLatest version NVIDIA Triton Inference Server release on October 2, 2024
Name | NVIDIA Triton Inference Server |
Category Name | Servers |
Release Date | October 2, 2024 |
LTS Version | 24.10 |
License | 0 |
Official Website | https://developer.nvidia.com/triton-inference-server |
Supported OS | Windows, Linux |
NVIDIA Triton Inference Server is a versatile open-source platform that streamlines AI model deployment across diverse environments, including cloud, data center, edge, and embedded devices. It supports multiple frameworks such as TensorFlow, PyTorch, ONNX, and OpenVINO, enabling concurrent model execution with features like dynamic batching and model ensembling. Triton offers both HTTP/REST and gRPC protocols for seamless integration into various applications.
Software Requirements:- Linux: Ubuntu 22.04 or later, Docker, NVIDIA Container Toolkit, and CUDA drivers compatible with CUDA 12.4. Windows: Windows 10 or later, Docker Desktop with WSL 2 backend, and NVIDIA GPU driver version 545 or later. Hardware Requirements:- Linux: NVIDIA GPU with compute capability 6.0 or higher (Pascal architecture or newer), sufficient CPU resources, and at least 16 GB RAM. Windows: NVIDIA GPU with compute capability 6.0 or higher, multi-core CPU, and a minimum of 16 GB RAM.
Latest version 1.0.0 release on January 10, 2024