NeuReality’s New Performance Results Pave the Way to Solve the World’s Growing AI Energy and Cost Crisis
18 Junio 2024 - 6:00AM
Business Wire
Real-World Performance on Common AI
Pipelines Show Up to 90% Cost Savings and 15x Better Energy
Efficiency Over Today’s AI Inference Servers
NeuReality, a leader in AI technology, has announced remarkable
performance results from its commercially available NR1-S™ AI
Inference Appliance, which significantly cuts costs and energy use
in AI data centers, offering a much-needed solution to the growing
concerns over AI’s high expenses and energy consumption. As
governments, environmental organizations, and businesses raise
alarms over AI’s unsustainable power consumption and exorbitant
costs, NeuReality's breakthrough comes at a critical time with the
explosive growth of generative AI. The NR1-S solution provides a
responsible and affordable option for the 65% of global and 75% of
U.S. businesses and governments struggling to adopt AI today.
The NR1-S does not compete with GPUs or other AI accelerators
but rather boosts their output and complements them. NeuReality’s
published results compare the NR1-S inference appliance paired with
Qualcomm® Cloud AI 100 Ultra and Pro accelerators against
traditional CPU-centric inference servers with Nvidia® H100 or L40S
GPUs. The NR1-S achieves dramatically improved cost savings and
energy efficiency in AI data centers across common AI applications
compared to the CPU-centric systems currently relied upon by
large-scale cloud service providers (hyperscalers), server OEMs and
manufacturers such as Nvidia.
Key Benefits from NR1-S
Performance
According to a technical blog shared on Medium this morning,
NeuReality’s real-world performance findings show the following
improvements:
- Massive Cost Savings: When paired with AI 100 Ultra,
NR1-S achieves up to 90% cost savings across various AI data types,
such as image, audio and text. These are the key building blocks
for generative AI, including large language models, mixture of
experts (MoE), retrieval-augmented generation (RAG) and
multimodality.
- Critical Energy Efficiency: Besides saving on the
capital expenditure (CAPEX) of AI use cases, the NR1-S shows up to
15 times better energy efficiency compared to traditional
CPU-centric systems, further reducing operational expenditure
(OPEX).
- Optimal AI Accelerator Use: Unlike traditional
CPU-centric systems, NR1-S ensures 100% utilization of the
integrated AI accelerators without performance drop-offs or delays
observed in today’s CPU-reliant systems.
Critical Impact for Ever-Evolving
Real-World AI Applications
The performance data included key metrics like AI queries per
dollar, queries per watt, and total cost of 1 million queries (both
CAPEX and OPEX). The data zone in on natural language processing
(NLP), automatic speech recognition (ASR), and computer vision (CV)
commonly used in medical imaging, fraud detection, customer call
centers, online assistants and much more:
- Cost Efficiency: One of the ASR tests shows NR1-S
cutting the cost of processing 1 million audio seconds from 43
cents to only 5 cents, making voice bots and other audio-based NLP
applications more affordable and capable of handling more
intelligence per query.
- Energy Savings: The tests also measured energy
consumption, with ASR showing seven seconds of audio processing per
watt with NR1-S, compared to 0.7 seconds in traditional CPU-centric
systems. This translates to a 10-fold increase in performance for
the energy used.
- Linear scalability: The NR1-S demonstrates the same
performance output regardless of the number of AI accelerators
used, allowing customers to efficiently scale their AI
infrastructure up or down with zero performance loss. This ensures
maximum return on investment without the diminishing returns
typically caused by adding more GPUs or other accelerators in
CPU-centric servers.
The NR1-S offers a practical solution for businesses and
governments looking to adopt AI without breaking the bank or
overloading power grids. It supports a variety of AI applications
commonly used in the financial services, healthcare, biotechnology,
entertainment, content creation, government, public safety and
transportation sectors.
These real-world performance results provide a welcome remedy to
the energy crisis facing AI infrastructure providers and
next-generation hyperscalers’ supercomputers. “While faster and
faster GPUs drive innovation in new AI capabilities, the current
systems that support them also move us further away from the budget
and carbon reduction goals of most companies,” said NeuReality
Chief R&D Officer Ilan Avital. “Our NR1-S is designed to
reverse that trend, enabling sustainable AI growth without
sacrificing performance.”
“As the industry keeps racing forward with a narrow focus on raw
performance for the biggest AI models, energy consumption and costs
keep skyrocketing,” said NeuReality co-founder and CEO Moshe
Tanach. “The NR1-S technology allows our customers to scale AI
applications affordably and sustainably, ensuring they can achieve
their business objectives and environmental targets. NeuReality was
built from inception to solve the cost and energy problem in AI
inferencing, and our new data clearly show we have developed a
viable solution. It’s an exciting step forward for the AI
industry.”
About NeuReality
Founded in 2019, NeuReality creates AI systems that are easy to
use, affordable, and energy efficient. Their goal is to make AI
accessible to everyone, from small businesses to large
governments.
Learn More: For detailed test results and more
information, visit the Full Blog.
View source
version on businesswire.com: https://www.businesswire.com/news/home/20240618910876/en/
Media Contact Treble Will Kruisbrink
neureality@treblepr.com