Rapt AI, a supplier of AI-powered AI-workload automation for GPUs and AI accelerators, has teamed with AMD to reinforce AI infrastructure.
The long-term strategic collaboration goals to enhance AI inference and coaching workload administration and efficiency on AMD Intuition GPUs, providing clients a scalable and cost-effective answer for deploying AI purposes.
As AI adoption accelerates, organizations are grappling with useful resource allocation, efficiency bottlenecks, and complicated GPU administration.
By integrating Rapt’s clever workload automation platform with AMD Intuition MI300X, MI325X and upcoming MI350 sequence GPUs, this collaboration delivers a scalable, high-performance, and cost-effective answer that allows clients to maximise AI inference and coaching effectivity throughout on-premises and multi-cloud infrastructures.
A extra environment friendly answer
Charlie Leeming, CEO of Rapt AI, mentioned in a press briefing, “The AI models we are seeing today are so large and most importantly are so dynamic and unpredictable. The older tools for optimizing don’t really fit at all. We observed these dynamics. Enterprises are throwing lots of money. Hiring a new set of talent in AI. It’s one of these disruptive technologies. We have a scenario where CFOs and CIOs are asking where is the return. In some cases, there is tens of millions, hundreds of millions or billions of dollars spend on GPU-related infrastructure.”
Leeming mentioned Anil Ravindranath, CTO of Rapt AI, noticed the answer. And that concerned deploying displays to allow observations of the infrastructure.
“We feel we have the right solution at the right time. We came out of stealth last fall. We are in a growing number of Fortune 100 companies. Two are running the code among cloud service providers,” Leeming mentioned.
And he mentioned, “We do have strategic partners but our conversations with AMD went extremely well. They are building tremendous GPUs, AI accelerators. We are known for putting the maximum amount of workload on GPUs. Inference is taking off. It’s in production stage now. AI workloads are exploding. Their data scientists are running as fast as they can. They are panicking, they need tools, they need efficiency, they need automation. It’s screaming for the right solution. Inefficiencies — 30% GPU underutilization. Customers do want flexibility. Large customers are asking if you support AMD.”
Enhancements that may take 9 hours could be finished in three minutes, he mentioned. Ravindranath mentioned in a press briefing the Rapt AI platform permits as much as 10 instances mannequin run capability on the similar AI compute spending degree, as much as 90% value financial savings, and nil people in a loop and no code modifications. For productiveness, this implies no extra ready for compute and time spent tuning infrastructure.
Lemming mentioned different methods have been round for some time and haven’t reduce it. Run AI, a rival, overlaps in a aggressive method considerably. He mentioned his firm observes in minutes as a substitute of hours after which optimizes the infrastructure. Ravindranath mentioned Run AI is extra like a scheduler however Rapt AI positions itself for unpredictable outcomes and offers with it.
“We run the model and figure it out, and that’s a huge benefit for inference workloads. It should just automatically run,” Ravindranath mentioned.
The advantages: decrease prices, higher GPU utilization
The businesses mentioned that AMD Intuition GPUs, with their industry-leading reminiscence capability, mixed with
Rapt’s clever useful resource optimization, helps guarantee most GPU utilization for AI workloads, serving to decrease complete value of possession (TCO).
Rapt’s platform streamlines GPU administration, eliminating the necessity for information scientists to spend invaluable time on trial-and-error infrastructure configurations. By robotically optimizing useful resource allocation for his or her particular workloads, it empowers them to deal with innovation somewhat than infrastructure. It seamlessly helps numerous GPU environments (AMD and others, whether or not within the cloud, on premises or each) via a single occasion, serving to guarantee most infrastructure flexibility.
The mixed answer intelligently optimizes job density and useful resource allocation on AMD Intuition GPUs, leading to higher inference efficiency and scalability for manufacturing AI deployments. Rapt’s auto-scaling capabilities additional assist guarantee environment friendly useful resource use primarily based on demand, decreasing latency and maximizing value effectivity.
Rapt’s platform works out-of-the-box with AMD Intuition GPUs, serving to guarantee speedy efficiency advantages. Ongoing collaboration between Rapt and AMD will drive additional optimizations in thrilling areas corresponding to GPU scheduling, reminiscence utilization and extra, serving to guarantee clients are geared up with a future prepared AI infrastructure.
“At AMD, we are committed to delivering high-performance, scalable AI solutions that empower organizations to unlock the full potential of their AI workloads.” mentioned Negin Oliver, company vp of enterprise growth for information heart GPU enterprise at AMD, in an announcement. “Our collaboration with Rapt AI combines the cutting-edge capabilities of AMD Instinct GPUs with Rapt’s intelligent workload automation, enabling customers to achieve greater efficiency, flexibility, and cost savings across their AI infrastructure.”
