Home Startups AI Chip maker Mythic Raises $40M to Develop a New Line AI-Based...

AI Chip maker Mythic Raises $40M to Develop a New Line AI-Based Hardware

Huge financial deals appear to be favoring AI companies nowadays. Mythic, an AI startup, recently joined the list of beneficiaries by raising $40 million in its recent round of funding led by SoftBank Ventures.

LockHeed Martin Ventures made a strategic investment in the company.

In addition, Reen Haass, ARM executive, is also set to join Mythic’s board of directors. The company concentrates on the inference part of AI activities, which involves undertaking calculations on the spot for things that are based off a broadly trained model.

Michael Henry, the startup’s CEO said that the key to getting excellent energy efficiency and high performance entails keeping everything on the chip.

He went ahead to say that going outside the chip to the memory would cause you to lose all the energy and performance. With that knowledge, Mythic realized that you could utilize flash memory uniquely.

However, the company has encountered challenges, especially in getting the memory and processors close together to avoid transferring the data around on the particular chip.

Just like other startups, Mythic seeks to ease the numerous trips made to the memory located on the processors in a bid to accelerate things while lowering the power consumption.

Nevertheless, Michael Henry revealed that Mythic had discovered how to carry out the operations, related to linear algebra, on flash memory.

The approach used by Mythic is designed to be more analog. The company’s chips are made to complete analog operations for both multiplication and addition in an attempt to deal with computational needs for an inference task.

According to Henry, the result dissipates less heat and consumes less power while receiving enough accuracy to obtain the ideal solution. The problem with the approach is adding a layer on top to make it appear and act like a normal chip.

Mythic’s goal involves plugging into various frameworks such as TensorFlow. The benefit of such frameworks is that they block out all the sophisticated tuning and tooling needed for such a piece of hardware and make it easy for developers to begin creating machine learning projects.

Andre Feldman, the CEO of Cerebras Systems, another AI hardware startup, asserted that frameworks such as TensorFlow possess most of the value that Nvidia had when creating an ecosystem for developers on its system.

Henry hopes that Mythic will surpass GPU’s huge performance gains. In fact, the company’s mission involves matching the performance of $1000 GPU while proving that it can consume less power and take up less space.

With the availability of a market for a chip that clients can swap immediately, Mythic aims at utilizing a PCI-E interface, a popular plug-and-play system. The only problem is getting into the actual design.

The competition between AI hardware startups is expected to rise, looking at the recent massive financing rounds in the industry. Big players such as Nvidia control a significant part of the market thanks to its GPUs being more efficient in performing the correct math for AI.

Henry distinguished Mythic from Nvidia by saying that both companies serve in different markets, whereby Mythic focuses on things that are embedded inside cars, phones, robotics and drones for uses like VR and AR.

Source TechCrunch

Subscribe to our newsletter

Signup today for free and be the first to get notified on the latest news and insights on artificial intelligence

KC Cheung
KC Cheung
KC Cheung has over 18 years experience in the technology industry including media, payments, and software and has a keen interest in artificial intelligence, machine learning, deep learning, neural networks and its applications in business. Over the years he has worked with some of the leading technology companies, building and growing dynamic teams in a fast moving international environment.
- Advertisment -


How can AI be leveraged for Clinical Trial Prediction?

Successful clinical trials occur at a ratio of 1 in 10, costing around $2-$3 billion, with drugs taking 10–12 years to be approved. High costs,...