The field of artificial intelligence continues to develop, and innovation in chips and software is always the key to promoting progress. Recently, the emerging AI company Recogni launched the revolutionary computing method Pareto, which is expected to completely change the landscape of AI chips. The Pareto system uses an innovative logarithmic method to significantly improve performance while reducing energy consumption and costs when running large AI models. This is supported by industry giants such as BMW and Bosch and well-known venture capital institutions. The editor of Downcodes will explain this technology and its potential impact in detail.
In the wave of artificial intelligence, innovation in chips and software has always been the key to promoting the development of the industry. Now, the emerging AI company Recogni has brought a revolutionary computing method, which not only makes AI chips smaller and faster, but also significantly reduces operating costs.
Recogni’s new technology, called Pareto, uses an innovative logarithmic method that enables it to surpass the performance indicators of existing technologies when running large AI models. This breakthrough is backed by well-known corporate and venture capital firms such as BMW, Bosch and Mayfield.
Traditional AI models, such as OpenAI’s GPT-4 and Google’s Gemini, require a lot of power-consuming calculations when processing simple chat prompts. Recogni's Pareto system significantly reduces energy consumption by converting multiplication operations into addition operations while maintaining model accuracy.
Picture source note: The picture is generated by AI, and the picture is authorized by the service provider Midjourney
Recogni has tested the Pareto system on AI models developed by companies such as Meta and Stability AI, with positive results. The company's first chip, designed and manufactured using Taiwan Semiconductor Manufacturing Co.'s seven-nanometer process, is currently working with an unnamed partner and plans to announce a broader rollout in the coming months.
Gilles Backhus, co-founder and vice president of AI at Recogni, said that the company is negotiating cooperation with some companies that provide data center hardware rental services, which will be an important direction for the deployment of Pareto technology.
Recogni’s Pareto system demonstrates exciting innovation potential in the field of AI chips. Its high efficiency and low cost are expected to promote the further development and popularization of artificial intelligence technology and provide strong impetus for more application scenarios. We look forward to Recogni’s broader promotion plans in the coming months and how Pareto technology will perform in practical applications. The editor of Downcodes will continue to pay attention to the latest developments in this field and bring more exciting reports to readers.