resnet34inf_b&w_inv_comp
 

Microsoft and Graphcore collaborate to accelerate Artificial Intelligence

The Microsoft IPU Cloud Preview on Azure - the world's first IPU cloud service is the first commercial hyperscale cloud to have launched with a new class of silicon processor built from the ground up to support the next generation of machine intelligence applications. 

Read the Full Story
Azure

The world's first IPU preview on Azure is now open 

Learn more
DellEMC

Introducing the world's first Graphcore IPU server 

Learn more
Bert

Check out our State of The Art Benchmarks

Learn more
Bert-1

Natural Language Processing - BERT

We have achieved state of the art performance and accuracy with the BERT language model, training BERT Base in 56 hours with seven C2 IPU-Processor PCIe cards, each with two IPUs, in an IPU Server system. With BERT inference, we see 3x higher throughput with over 20% improvement in latency to serve up results faster than ever.

ResNext (1)

Image Classification - ResNext

Graphcore C2 IPU-Processor PCIe card achieves 3.4x higher throughput at 20x lower latency compared to a leading alternative processor. High throughput at the lowest possible latency is key in many of the important use cases today.

Nov 19, 2019
Graphcore announces bare-metal IPU cloud service with Cirrascale
Learn more
Nov 13, 2019
Microsoft and Graphcore collaborate to accelerate Artificial Intelligence
Learn more
Nov 13, 2019
Graphcore delivers 26x performance gains for Finance customers
Learn more
Nov 13, 2019
Probabilistic Modelling by Combining MCMC and Variational Inference with IPUs
Learn more

Microsoft Sends a New Kind of AI Processor Into the Cloud

Learn more

Graphcore's Plans to Disrupt Computer Processor Market (video interview)

Learn more

In Fierce AI Race, Everybody Wants This New Human-Like Chip

Learn more

Get the latest Graphcore news