AI benchmarking group, MLCommons, examines speed of responding to consumer queries

0
40
AI benchmarking group, MLCommons, examines speed of responding to consumer queries
AI benchmarking group, MLCommons, examines speed of responding to consumer queries

The artificial intelligence benchmarking group MLCommons unveiled a new set of tests and results that assess how quickly top-tier technology can run AI apps and respond to customers

The artificial intelligence benchmarking group MLCommons released a new set of tests and results that measure the speed at which top-tier technology can execute AI apps and reply to consumers on Wednesday.

MLCommons has added two new benchmarks that test the speed with which AI chips and systems can create answers from sophisticated AI models loaded with data. The findings generally represent how soon an AI application like ChatGPT can respond to a user query.

One of the new benchmarks allows you to test the speed of a question-and-answer scenario for large language models. Meta Platforms (META.O) produced Llama 2, which has 70 billion parameters and opens a new tab.

MLCommons administrators have now introduced a second text-to-image generator to the array of benchmarking tools, MLPerf, which is based on Stability AI’s Stable Diffusion XL model.

Servers powered by Nvidia’s H100 chips, built by the likes of Alphabet’s Google (GOOGL.O), Supermicro (SMCI.O), and Nvidia (NVDA.O), easily won both new raw performance benchmarks. Several server builders submitted designs using the company’s less powerful L40S processor.

Krai, a server maker, submitted a design for the image generation test using a Qualcomm (QCOM.O.) AI chip that uses far less power than Nvidia’s cutting-edge processors.

Intel (INTC.O.) also submitted a proposal using its Gaudi2 accelerator processors. The company described the outcomes as “solid.”.

Raw performance is not the only metric to consider when deploying AI systems. Advanced AI processors consume tremendous quantities of energy, and one of the most fundamental issues for AI firms is installing chips that provide the best performance for the least amount of energy.

MLCommons includes a second benchmark category for measuring power usage.

Also readNurturing Responsible Online Behavior in Students by Building a Culture of Digital Citizenship

Do FollowCIO News LinkedIn Account | CIO News Facebook | CIO News Youtube | CIO News Twitter 

About us:

CIO News, a proprietary of Mercadeo, produces award-winning content and resources for IT leaders across any industry through print articles and recorded video interviews on topics in the technology sector such as Digital Transformation, Artificial Intelligence (AI), Machine Learning (ML), Cloud, Robotics, Cyber-security, Data, Analytics, SOC, SASE, among other technology topics.