Top
Author
Published
1 Apr 2026Form Number
LP2413PDF size
5 pages, 504 KBSubscribed to LP2413.
Thank you for your feedback.
Abstract
As part of the MLCommons® MLPerf™ Inference v6.0 benchmark, Lenovo continues its commitment to advancing high-performance, enterprise-grade AI infrastructure. As an early and consistent participant in MLPerf, Lenovo reinforces its leadership by delivering scalable, versatile platforms designed to support the most demanding AI inference workloads—accelerated by Lenovo and NVIDIA’s AI infrastructure.
Introduction
Running today’s most advanced AI models requires processing massive volumes of data with extreme efficiency. That performance depends on a tightly optimized combination of compute, networking, storage, and cooling—engineered for reliability, resilience, and scale.
Lenovo and NVIDIA have worked together to deliver trusted AI infrastructure for enterprise and research environments. In MLPerf Inference v6.0, Lenovo again demonstrated powerful performance in accelerated computing across a broad set of industry-standard workloads, including large language models (LLMs), and graph neural networks.
Systems submitted were the following:
- Lenovo ThinkSystem SR680a V4 with 8× NVIDIA HGX B300-SXM (8x NVIDIA Blackwell Ultra) 2.1TB)
- Lenovo ThinkSystem SR675i V3 with 8× NVIDIA RTX™ PRO 6000 Blackwell Server Edition
Figure 1. Lenovo ThinkSystem SR680a V4 includes 8× NVIDIA B300-SXM GPUs
Highlights of MLPerf Inference v6.0
The Lenovo ThinkSystem SR680a V4 is next-gen, ultra-dense AI platform engineered for industrialized AI factories. Designed with Lenovo Neptune™ liquid-cooling technology, it delivers GPU-dense compute optimized for training, refining, and deploying large-scale AI models across a diverse range of workloads, proving itself with strong performance in LLMs and multimodal tasks.
The Lenovo ThinkSystem SR675i V3 is the GPU-dense platform that is optimized for AI inference and powerful enough for fine-tuning, and simulation across verticals, delivering exceptional parallelism, memory bandwidth, and throughput. GPUs receive data without disruption to maximize utilization. Engineered for AI factories at scale, enterprises can realize better cost/token, faster ROI, and accelerated training for larger models.
Benchmarks submitted:
- LLM llama2-70b
- GPT-OSS-120B
- Automatic Speech Recognition – whisper
Figure 2. ThinkSystem SR675i V3 Inference Model includes 8× NVIDIA RTX PRO 6000 Blackwell Server Edition GPUs
Empowering Businesses with Lenovo’s AI Leadership
Lenovo’s MLPerf benchmark results reinforce our continued focus on delivering high-performance, future-ready AI systems that meet the evolving needs of modern workloads.
Key differentiators include:
- Cross-Workload Excellence: Lenovo systems consistently deliver strong results across LLM use cases.
- Scalable Infrastructure: Built with the industry’s first AI optimized inference server powered by NVIDIA RTX PRO 6000 Blackwell Server Edition GPUs, NVIDIA HGX B300 Blackwell Ultra for large-scale AI training, fine-tuning and inference. Lenovo’s systems offer the performance and flexibility needed for AI projects of any size.
- Enterprise Integration: With support for massive inference at scale, centralized training and Hybrid AI deployment, Lenovo empowers organizations to scale AI from research labs to production environments.
Looking Ahead: AI Innovation with Lenovo & NVIDIA
As AI workloads continue to grow in complexity and scale, Lenovo remains firmly committed to innovation. Our strong MLPerf Inference results reflect not only technical excellence but also a forward-thinking approach to solving tomorrow’s AI challenges with real business outcomes.
With deep industry collaborations like NVIDIA and a focus on engineering co-designed agile, efficient infrastructure, Lenovo carves a leading path for breakthroughs in generative AI, LLMs, AI Agents, Recommender Systems, and more.
Conclusion
Lenovo’s latest MLPerf Inference results reaffirm its leadership in advanced AI infrastructure. Powered by ThinkSystem SR675i V3 and SR680a V4 servers, Lenovo delivers powerful performance, scalability, and unmatched reliability—whether training complex language models or enabling real-time inference for generative and agentic AI. For organizations ready to lead in AI, Lenovo Hybrid AI Advantage™ with NVIDIA provides a trusted foundation to build, train, and scale with confidence—unlocking innovation and accelerating Smarter AI.
Lenovo and NVIDIA are driving the future of AI with faster, Smarter AI for All.
Read the NVIDIA Extreme co-design delivers new mlperf inference records blog or
See more NVIDIA performance data on the Data Center Deep Learning Product Performance Hub and Performance Explorer pages.
Author
Traci Parker is the Worldwide Solutions Marketing Manager for Enterprise IT and AI at Lenovo. She specializes in hybrid cloud, infrastructure modernization and AI solutions. She has more than 15 years of experience as a Marketing Manager and Product Marketing Manager across high-tech, fin-tech and healthcare industries.
Trademarks
Lenovo and the Lenovo logo are trademarks or registered trademarks of Lenovo in the United States, other countries, or both. A current list of Lenovo trademarks is available on the Web at https://www.lenovo.com/us/en/legal/copytrade/.
The following terms are trademarks of Lenovo in the United States, other countries, or both:
Lenovo®
Lenovo Hybrid AI Advantage
Neptune®
ThinkSystem®
Other company, product, or service names may be trademarks or service marks of others.
Configure and Buy
Please select a locale
Full Change History
Course Detail
Employees Only Content
The content in this document with a is only visible to employees who are logged in. Logon using your Lenovo ITcode and password via Lenovo single-signon (SSO).
The author of the document has determined that this content is classified as Lenovo Internal and should not be normally be made available to people who are not employees or contractors. This includes partners, customers, and competitors. The reasons may vary and you should reach out to the authors of the document for clarification, if needed. Be cautious about sharing this content with others as it may contain sensitive information.
Any visitor to the Lenovo Press web site who is not logged on will not be able to see this employee-only content. This content is excluded from search engine indexes and will not appear in any search results.
For all users, including logged-in employees, this employee-only content does not appear in the PDF version of this document.
This functionality is cookie based. The web site will normally remember your login state between browser sessions, however, if you clear cookies at the end of a session or work in an Incognito/Private browser window, then you will need to log in each time.
If you have any questions about this feature of the Lenovo Press web, please email David Watts at dwatts@lenovo.com.
