AI chip startup Blaize nabs $106M in fresh funding

Startup Blaize Inc., which supplies artificial intelligence chips for edge devices and data centers, today disclosed that it has raised $106 million in fresh funding.

The capital was provided by a consortium that included Mercedes Benz, Franklin Templeton and a half-dozen other backers. The round comes lower than six months after Blaize announced plans to go public through a merger with a special purpose acquisition company, or SPAC. The chipmaker said on the time that the deal was expected to lift a minimum of $71 million.

Most AI models are implemented as a group of code snippets called artificial neurons. Each such code snippet performs a small portion of the duty assigned to the AI model wherein it runs. Blaize has developed a system-on-chip, the Blaize 1600, that it says can run neural networks with higher power efficiency than rival products.

At the guts of the chip’s design is a pc science concept referred to as graph processing. A graph is an information structure that incorporates multiple pieces of data dubbed nodes. Related pieces of data are linked together by connections, or edges, that help explain the connection between them.

Practically all AI models could be represented as a graph. A model’s artificial neurons could be encoded into the nodes, or information snippets, that make up a graph. Meanwhile, the style wherein those neurons work together to crunch data could also be represented using the connections that link together the nodes.

Blaize developed its Blaze 1600 chip from the bottom as much as process graphs. Based on the corporate, the processor’s specialized design allows it to run AI models using less energy than graphics cards or field-programmable gate arrays. That makes it well-suited for powering connected devices, which frequently have battery life constraints. 

The Blaze 1600 features 16 cores optimized for graph processing. It also includes optimizations designed to hurry up the duty of processing intermediate results, additional information that AI models generate while crunching input data. Blaize says the chip can manage as much as 16 trillion computing operations per second.

“Our unique, fully programmable approach makes us ready for the unknown,” said Blaize Chief Executive Officer Dinakar Munagala. “This is good within the fast-changing AI applications landscape, de-risking and reducing cost for our customers, scaling from the sting to the info center, with one uniform and complete hardware and software solution.”

The corporate sells its silicon in multiple form aspects. For the sting computing market, Blaize offers compute modules that mix the Blaize 1600 with supporting components comparable to memory. Those modules have a compact design that makes it relatively easy to integrate them into connected devices.

The corporate also targets its chip at data centers. It offers a PCIe accelerator, the Xplorer X1600P-Q, with a single Blaize 1600 that customers can attach to existing servers. For organizations with more advanced requirements, the chipmaker offers a server of its own that mixes 24 of its processors with two Intel Corp. central processing units.

Blaize will use the capital from its newly announced funding round to finance product development and commercialization initiatives. Based on the corporate, the hassle will place an emphasis on expanding its presence within the automotive, computer vision, AI inference and generative AI segments.

Image: Blaize