AI startup Recogni unveils new computing method to slash costs, power requirements
By Akash Sriram
Aug 20 (Reuters) -AI chip and software startup Recogni unveiled a novel computing method on Tuesday that could make its chips used to train and run artificial intelligence systems smaller, faster and less expensive to operate.
Backed by BMW BMWG.DE, Bosch and venture capital firm Mayfield, Recogni develops specialized chips and software to enable AI inferencing - the process of trained AI models making predictions or decisions on new, unseen data.
The company said the new patented system, called Pareto, utilizes a logarithmic approach that outperforms existing methods when running large AI models.
"It is a huge leap in all of the KPIs (key performance indicators) that influence silicon hardware system design when it comes to AI computing," Recogni's co-founder and VP of AI, Gilles Backhus told Reuters.
Current AI models, such as OpenAI's GPT-4 and Google's Gemini, require hundreds of thousands of power-hungry mathematical operations for the simple of prompts on chatbots like ChatGPT.
Recogni said that its new system converts these multiplication operations into additions, significantly reducing power consumption while maintaining accuracy.
The startup said it has already tested Pareto on AI models developed by Meta Platforms, Stability AI, and others.
Recogni, whose first chip was designed, manufactured Taiwan Semiconductor Manufacturing Co's 2330.TW seven nanometer process, said it was working with an unnamed partner to make Pareto more widely available and will announce the partnership in the coming months.
"We are speaking to companies that are putting hardware in data centers and offering it to the world to whoever wants to basically rent it ... that's definitely one of the deployment routes that we're considering," Backhus added.
Reporting by Akash Sriram in Bengaluru; Editing by Tasim Zahid
Related Assets
Latest News
Disclaimer: The XM Group entities provide execution-only service and access to our Online Trading Facility, permitting a person to view and/or use the content available on or via the website, is not intended to change or expand on this, nor does it change or expand on this. Such access and use are always subject to: (i) Terms and Conditions; (ii) Risk Warnings; and (iii) Full Disclaimer. Such content is therefore provided as no more than general information. Particularly, please be aware that the contents of our Online Trading Facility are neither a solicitation, nor an offer to enter any transactions on the financial markets. Trading on any financial market involves a significant level of risk to your capital.
All material published on our Online Trading Facility is intended for educational/informational purposes only, and does not contain – nor should it be considered as containing – financial, investment tax or trading advice and recommendations; or a record of our trading prices; or an offer of, or solicitation for, a transaction in any financial instruments; or unsolicited financial promotions to you.
Any third-party content, as well as content prepared by XM, such as: opinions, news, research, analyses, prices and other information or links to third-party sites contained on this website are provided on an “as-is” basis, as general market commentary, and do not constitute investment advice. To the extent that any content is construed as investment research, you must note and accept that the content was not intended to and has not been prepared in accordance with legal requirements designed to promote the independence of investment research and as such, it would be considered as marketing communication under the relevant laws and regulations. Please ensure that you have read and understood our Notification on Non-Independent Investment. Research and Risk Warning concerning the foregoing information, which can be accessed here.
