Amazon Nova Canvas, Anthropic, and Nova Reel: A State of the Art and a Challenge for AI Intelligence Platforms

Amazon announced these new models at its AWS re:Invent conference, which is happening now in Las Vegas. At the show, the company also stated that it’s collaborating with Anthropic to build a humongous Artificial Intelligence compute cluster that will be powered by its Trainium 2 chips. “When completed, it is expected to be the world’s largest AI compute cluster reported to date available for Anthropic to build and deploy its future models on,” according to Amazon.

Amazon Nova Canvas, an image generation model and Amazon Nova Reel are two of the content generation models the company is releasing. The company says that it has watermarking capabilities for these models. As an example of what’s possible with Nova Reel, Amazon has shared this mock ad for a fake pasta brand.

The company is racing to release new products and features to stay ahead of other companies. Large companies may be more willing to use Amazon’s artificial intelligence offerings because of the company’s reputation. An Apple exec even appeared onstage today at re:Invent to talk about how the company relies on Amazon’s custom AI chips.

The launch of the voice assistant will not happen this fall as was reported, but it will happen next year.

At the company’s Re:Invent conference in Vegas, Matt Garman, the CEO of Amazon Web Services, said that his company planned to make the world’s first machine learning system with supercomputing power.

Garman announced that the Tranium 2 will be made generally accessible in Trn2 Ultraserver clusters. Many companies already use Amazon’s cloud to build and train custom AI models, often in tandem with GPUs from Nvidia. But Garman said that the new AWS clusters are 30 to 40 percent cheaper than those that feature Nvidia’s GPUs.

Patrick Moorhead, CEO and the chief analyst at Moore Insight & Strategy said the numbers were astounding for the next-Generation chip. Moorhead says that Trainium 3 appears to have received a significant performance boost from an improvement in the so-called interconnect between chips. The ability to quickly transfer data between chips is a factor that can be critical in the development of large Artificial Intelligence models.

Moorehead predicts that the competition will increase over the next several years and that Nvidia will face more competition in the future. He says Amazon is showing that there is more than one game for training.