PanGu-Σ

PanGu-Σ

Huawei
SYCL

SYCL

The Khronos Group
+
+

Related Products

  • LM-Kit.NET
    26 Ratings
    Visit Website
  • Vertex AI
    961 Ratings
    Visit Website
  • Google AI Studio
    11 Ratings
    Visit Website
  • Ango Hub
    15 Ratings
    Visit Website
  • Dragonfly
    16 Ratings
    Visit Website
  • Dynamo Software
    68 Ratings
    Visit Website
  • Datasite Diligence Virtual Data Room
    640 Ratings
    Visit Website
  • Google Cloud BigQuery
    2,008 Ratings
    Visit Website
  • LTX
    181 Ratings
    Visit Website
  • Concord
    237 Ratings
    Visit Website

About

Significant advancements in the field of natural language processing, understanding, and generation have been achieved through the expansion of large language models. This study introduces a system which utilizes Ascend 910 AI processors and the MindSpore framework to train a language model with over a trillion parameters, specifically 1.085T, named PanGu-{\Sigma}. This model, which builds upon the foundation laid by PanGu-{\alpha}, takes the traditionally dense Transformer model and transforms it into a sparse one using a concept known as Random Routed Experts (RRE). The model was efficiently trained on a dataset of 329 billion tokens using a technique called Expert Computation and Storage Separation (ECSS), leading to a 6.3-fold increase in training throughput via heterogeneous computing. Experimentation indicates that PanGu-{\Sigma} sets a new standard in zero-shot learning for various downstream Chinese NLP tasks.

About

SYCL is an open, royalty-free, cross-platform programming standard defined by the Khronos Group that enables heterogeneous and offload computing in modern ISO C++ by providing a single-source abstraction layer where host and device code coexist in the same C++ source and can target a wide range of devices such as CPUs, GPUs, FPGAs, and other accelerators. It is a C++ API and abstraction designed to make heterogeneous computing more productive and portable by using standard language features such as templates, inheritance, and lambda expressions so developers can write and manage data and execution across diverse hardware without resorting to proprietary languages or extensions. SYCL builds on concepts of underlying acceleration backends like OpenCL and allows integration with other technologies while providing a consistent language, APIs, and ecosystem for locating devices, managing data, and executing kernels.

Platforms Supported

Windows
Mac
Linux
Cloud
On-Premises
iPhone
iPad
Android
Chromebook

Platforms Supported

Windows
Mac
Linux
Cloud
On-Premises
iPhone
iPad
Android
Chromebook

Audience

AI developers

Audience

Developers and engineers building high-performance applications for heterogeneous hardware who need a standard C++ programming model for parallel and accelerator computing

Support

Phone Support
24/7 Live Support
Online

Support

Phone Support
24/7 Live Support
Online

API

Offers API

API

Offers API

Screenshots and Videos

No images available

Screenshots and Videos

Pricing

No information available.
Free Version
Free Trial

Pricing

No information available.
Free Version
Free Trial

Reviews/Ratings

Overall 0.0 / 5
ease 0.0 / 5
features 0.0 / 5
design 0.0 / 5
support 0.0 / 5

This software hasn't been reviewed yet. Be the first to provide a review:

Review this Software

Reviews/Ratings

Overall 0.0 / 5
ease 0.0 / 5
features 0.0 / 5
design 0.0 / 5
support 0.0 / 5

This software hasn't been reviewed yet. Be the first to provide a review:

Review this Software

Training

Documentation
Webinars
Live Online
In Person

Training

Documentation
Webinars
Live Online
In Person

Company Information

Huawei
Founded: 1987
China
huawei.com

Company Information

The Khronos Group
Founded: 2000
United States
www.khronos.org

Alternatives

LTM-1

LTM-1

Magic AI

Alternatives

OpenCL

OpenCL

The Khronos Group
PanGu-α

PanGu-α

Huawei
DeepSeek-V2

DeepSeek-V2

DeepSeek
VideoPoet

VideoPoet

Google
OPT

OPT

Meta

Categories

Categories

Integrations

C++
PanGu Chat

Integrations

C++
PanGu Chat
Claim PanGu-Σ and update features and information
Claim PanGu-Σ and update features and information
Claim SYCL and update features and information
Claim SYCL and update features and information