Top
Author
Updated
19 Feb 2026Form Number
LP1732PDF size
25 pages, 2.3 MBSubscribed to LP1732.
Thank you for your feedback.
Abstract
The ThinkSystem NVIDIA H100 GPU delivers unprecedented performance, scalability, and security for every workload. The GPUs use breakthrough innovations in the NVIDIA Hopper™ architecture to deliver industry-leading conversational AI, speeding up large language models by 30X over the previous generation.
This product guide provides essential presales information to understand the NVIDIA H100 GPU and their key features, specifications, and compatibility. This guide is intended for technical specialists, sales specialists, sales engineers, IT architects, and other IT professionals who want to learn more about the GPUs and consider their use in IT solutions.
Change History
Changes in the February 19, 2026 update:
- Feature C1HL (withdrawn) is air-cooled, not water cooled - Technical specifications section
Introduction
The ThinkSystem NVIDIA H100 GPU delivers unprecedented performance, scalability, and security for every workload. The GPUs use breakthrough innovations in the NVIDIA Hopper™ architecture to deliver industry-leading conversational AI, speeding up large language models by 30X over the previous generation.
The NVIDIA H100 GPU features fourth-generation Tensor Cores and the Transformer Engine with FP8 precision, further extending NVIDIA’s market-leading AI leadership with up to 9X faster training and an incredible 30X inference speedup on large language models. For high-performance computing (HPC) applications, The GPUs triple the floating-point operations per second (FLOPS) of FP64 and add dynamic programming (DPX) instructions to deliver up to 7X higher performance.
The following figure shows the ThinkSystem NVIDIA H100 GPU in the double-width PCIe adapter form factor.
Figure 1. ThinkSystem NVIDIA H100 NVL 94GB PCIe Gen5 Passive GPU
Did you know?
The NVIDIA H100 family is available in both double-wide PCIe adapter form factor and in SXM form factor. The latter is used in Lenovo's Neptune direct-water-cooled ThinkSystem SD665-N V3 server for the ultimate in GPU performance and heat management.
The NVIDIA H100 NVL Tensor Core GPU is optimized for Large Language Model (LLM) Inferences, with its high compute density, high memory bandwidth, high energy efficiency, and unique NVLink architecture.
Part number information
The following table shows the part numbers for the ThinkSystem NVIDIA H100 GPU.
Not available in China, Hong Kong and Macau: The H100 GPUs are not available in China, Hong Kong and Macau. For these markets, the H800 is avalable. See the NVIDIA H800 product guide for details, https://lenovopress.lenovo.com/LP1814.
The PCIe option part numbers includes the following:
- One GPU with full-height (3U) adapter bracket attached
- Documentation
The following figure shows the NVIDIA H100 SXM5 8-GPU Board with heatsinks installed in the ThinkSystem SR680a V3 and ThinkSystem SR685a V3 servers.
Figure 2. NVIDIA H100 SXM5 8-GPU Board in the ThinkSystem SR680a V3 and SR685a V3 servers
Features
The ThinkSystem NVIDIA H100 GPU delivers high performance, scalability, and security for every workload. The GPU uses breakthrough innovations in the NVIDIA Hopper™ architecture to deliver industry-leading conversational AI, speeding up large language models (LLMs) by 30X over the previous generation.
The PCIe versions of the NVIDIA H100 GPUs include a five-year software subscription, with enterprise support, to the NVIDIA AI Enterprise software suite, simplifying AI adoption with the highest performance. This ensures organizations have access to the AI frameworks and tools they need to build accelerated AI workflows such as AI chatbots, recommendation engines, vision AI, and more.
The NVIDIA H100 GPU features fourth-generation Tensor Cores and the Transformer Engine with FP8 precision, further extending NVIDIA’s market-leading AI leadership with up to 9X faster training and an incredible 30X inference speedup on large language models. For high-performance computing (HPC) applications, the GPU triples the floating-point operations per second (FLOPS) of FP64 and adds dynamic programming (DPX) instructions to deliver up to 7X higher performance. With second-generation Multi-Instance GPU (MIG), built-in NVIDIA confidential computing, and NVIDIA NVLink Switch System, the NVIDIA H100 GPU securely accelerates all workloads for every data center from enterprise to exascale.
Key features of the NVIDIA H100 GPU:
- NVIDIA H100 Tensor Core GPU
Built with 80 billion transistors using a cutting-edge TSMC 4N process custom tailored for NVIDIA’s accelerated compute needs, H100 is the world’s most advanced chip ever built. It features major advances to accelerate AI, HPC, memory bandwidth, interconnect, and communication at data center scale.
- Transformer Engine
The Transformer Engine uses software and Hopper Tensor Core technology designed to accelerate training for models built from the world’s most important AI model building block, the transformer. Hopper Tensor Cores can apply mixed FP8 and FP16 precisions to dramatically accelerate AI calculations for transformers.
- NVLink Switch System
The NVLink Switch System enables the scaling of multi-GPU input/output (IO) across multiple servers. The system delivers up to 9X higher bandwidth than InfiniBand HDR on the NVIDIA Ampere architecture.
- NVIDIA Confidential Computing
NVIDIA Confidential Computing is a built-in security feature of Hopper that makes NVIDIA H100 the world’s first accelerator with confidential computing capabilities. Users can protect the confidentiality and integrity of their data and applications in use while accessing the unsurpassed acceleration of H100 GPUs.
- Second-Generation Multi-Instance GPU (MIG)
The Hopper architecture’s second-generation MIG supports multi-tenant, multi-user configurations in virtualized environments, securely partitioning the GPU into isolated, right-size instances to maximize quality of service (QoS) for 7X more secured tenants.
- DPX Instructions
Hopper’s DPX instructions accelerate dynamic programming algorithms by 40X compared to CPUs and 7X compared to NVIDIA Ampere architecture GPUs. This leads to dramatically faster times in disease diagnosis, real-time routing optimizations, and graph analytics.
The following figure shows the NVIDIA H100 SXM5 4-GPU Board installed in the ThinkSystem SD665-N V3 server
Figure 3. NVIDIA H100 SXM5 4-GPU Board in the ThinkSystem SD665-N V3 server
Technical specifications
The following table lists the GPU processing specifications and performance of the NVIDIA H100 GPU.
* With structural sparsity enabled
Server support
The following tables list the ThinkSystem servers that are compatible.
NVLink server support: The NVLink Ampere bridge is supported with additional NVIDIA A-series and H-series GPUs. As a result, there are additional servers listed as supporting the bridge that don't support the H100 GPU.
- Contains 8 separate GPUs connected via high-speed interconnects
- Contains 4 separate GPUs connected via high-speed interconnects
Operating system support
The following table lists the supported operating systems.
Tip: These tables are automatically generated based on data from Lenovo ServerProven.
1 HW is not supported with EPYC 7002 processors.
2 For limitation, please refer Support Tip TT1064
3 For limitation, please refer Support Tip TT1591
4 Ubuntu 22.04.3 LTS/Ubuntu 22.04.4 LTS
NVIDIA GPU software
This section lists the NVIDIA software that is available from Lenovo.
The PCIe adapter H100 GPUs include a five-year software subscription, including enterprise support, to the NVIDIA AI Enterprise software suite:
- ThinkSystem NVIDIA H100 NVL 94GB PCIe Gen5 Passive GPU, 4X67A89325
- ThinkSystem NVIDIA H100 80GB PCIe Gen5 Passive GPU, 4X67A82257
This license is equivalent to part number 7S02001HWW listed in the NVIDIA AI Enterprise Software section below.
To activate the NVIDIA AI Enterprise license, see the following page:
https://www.nvidia.com/en-us/data-center/activate-license/
SXM GPUs: The NVIDIA AI Enterprise software suite is not included with the SXM H100 GPUs and will need to ordered separately if needed.
NVIDIA Enterprise Software
Lenovo offers the NVIDIA Enterprise cloud-native enterprise software, comprised of NVIDIA AI Enterprise (NVAIE) and NVIDIA Omniverse Enterprise. NVIDIA Enterprise is an end-to-end, cloud-native suite of AI and data analytics software, optimized, certified, and supported by NVIDIA to run on VMware vSphere and bare-metal with NVIDIA-Certified Systems™. It includes key enabling technologies from NVIDIA for rapid deployment, management, and scaling of AI workloads in the modern hybrid cloud.
NVIDIA Enterprise is licensed on a per-GPU basis. NVIDIA Enterprise products can be purchased as either a perpetual license with support services, or as an annual or multi-year subscription.
- The perpetual license provides the right to use the NVIDIA Enterprise software indefinitely, with no expiration. NVIDIA Enterprise with perpetual licenses must be purchased in conjunction with one-year, three-year, or five-year support services. A one-year support service is also available for renewals.
- The subscription offerings are an affordable option to allow IT departments to better manage the flexibility of license volumes. NVIDIA Enterprise software products with subscription includes support services for the duration of the software’s subscription license
The following table lists NVIDIA Enterprise license part numbers and feature codes.
For more information, see the NVIDIA Enterprise Sizing Guide.
NVIDIA HPC Compiler Software
Auxiliary power cables
The power cables needed for the H100 SXM GPUs are included with the supported servers.
The H100 PCIe GPU option part number does not ship with auxiliary power cables. Cables are server-specific due to length requirements. For CTO orders, auxiliary power cables are derived by the configurator. For field upgrades, cables will need to be ordered separately as listed in the table below.
| Auxiliary power cable needed with the SR650 V3, SR655 V3, SR665 V3, SR665, SR650 V2 |
|
* The option part numbers are for thermal kits and include other components needed to install the GPU. See the SR650 V3 product guide or SR655 V3 product guide or SR665 V3 product guide for details. |
| Auxiliary power cable needed with the SR675 V3 |
.png "Click to show a larger version of the image") 235mm 16-pin (2x6+4) cableOption: 4X97A84510, ThinkSystem SR675 V3 Supplemental Power Cable for H100 GPU Option Feature: BSD2 SBB: SBB7A65299 Base: SC17B39301 FRU: 03LE554 |
| Auxiliary power cable needed with the SR850 V3, SR860 V3 |
.png "Click to show a larger version of the image") 200mm 16-pin (2x6+4) cableOption: 4X97A88016, ThinkSystem SR850 V3/SR860 V3 H100 GPU Power Cable Option Kit Feature: BW28 SBB: SBB7A72759 Base: SC17B40604 FRU: 03LF915 |
| Auxiliary power cable needed with the SR670 V2 |
 215mm 16-pin (2x6+4) cableOption: 4X97A85027, ThinkSystem SR670 V2 H100/L40 GPU Option Power Cable Feature: BRWL SBB: SBB7A66339 Base: SC17B33046 FRU: 03KM845 |
.png "Click to show a larger version of the image")
Regulatory approvals
The NVIDIA H100 GPU has the following regulatory approvals:
- RCM
- BSMI
- CE
- FCC
- ICES
- KCC
- cUL, UL
- VCCI
Operating environment
The NVIDIA H100 GPU has the following operating characteristics:
- Ambient temperature
- Operational: 0°C to 50°C (-5°C to 55°C for short term*)
- Storage: -40°C to 75°C
- Relative humidity:
- Operational: 5-85% (5-93% short term*)
- Storage: 5-95%
* A period not more than 96 hours consecutive, not to exceed 15 days per year.
Warranty
One year limited warranty. When installed in a Lenovo server, the GPU assumes the server’s base warranty and any warranty upgrades.
Related publications
For more information, refer to these documents:
- ThinkSystem and ThinkAgile GPU Summary:
https://lenovopress.lenovo.com/lp0768-thinksystem-thinkagile-gpu-summary - ServerProven compatibility:
https://serverproven.lenovo.com/ - NVIDIA H100 product page:
https://www.nvidia.com/en-us/data-center/h100/ - NVIDIA Hopper Architecture page
https://www.nvidia.com/en-us/data-center/technologies/hopper-architecture/ - ThinkSystem SD665-N V3 product guide
https://lenovopress.lenovo.com/lp1613-thinksystem-sd665-n-v3-server - ThinkSystem SR680a V3 product guide
https://lenovopress.lenovo.com/lp1909-thinksystem-sr680a-v3-server - ThinkSystem SR685a V3 product guide
https://lenovopress.lenovo.com/lp1910-thinksystem-sr685a-v3-server
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®
Neptune®
ServerProven®
ThinkAgile®
ThinkSystem®
The following terms are trademarks of other companies:
AMD is a trademark of Advanced Micro Devices, Inc.
Intel®, the Intel logo and Xeon® are trademarks of Intel Corporation or its subsidiaries.
Linux® is the trademark of Linus Torvalds in the U.S. and other countries.
Microsoft, Windows, and Windows Server are trademarks of Microsoft Corporation in the United States, other countries, or both.
NVIDIA®, CUDA®, NVIDIA HGX®, NVIDIA Hopper, NVIDIA Omniverse®, NVIDIA RTX®, NVLink®, and OpenACC® are trademarks of NVIDIA Corporation.
Other company, product, or service names may be trademarks or service marks of others.
Configure and Buy
Please select a locale
Full Change History
Changes in the February 19, 2026 update:
- Feature C1HL (withdrawn) is air-cooled, not water cooled - Technical specifications section
Changes in the December 16, 2024 update:
- Removed the vGPU and Omniverse software part numbers as not supported with the H100 GPUs - NVIDIA GPU software section
Changes in the August 9, 2024 update:
- The following GPU is withdrawn from marketing:
- ThinkSystem NVIDIA H100 80GB PCIe Gen5 Passive GPU, 4X67A82257
Changes in the May 2, 2024 update:
- Corrections to the number of CUDA cores and Tensor cores - Technical specifications section
Changes in the April 23, 2024 update:
- Added the following 8-GPU offering:
- ThinkSystem NVIDIA HGX H100 80GB 700W 8-GPU Board, C1HL
Changes in the November 8, 2023 update:
- Clarified that the H100 GPUs in the PCIe form factor include a 5-year subscription to the NVIDIA AI Enterprise software suite - NVIDIA GPU software section
Changes in the September 12, 2023 update:
- The NVIDIA H800 GPUs are now covered in a separate product guide, https://lenovopress.lenovo.com/lp1814
Changes in the August 29, 2023 update:
- The following servers now support the H100 PCIe adapter - Server support section
- SR665, SR650 V2, SR670 V2
- Added the power cables for the new servers - Auxiliary power cables section:
Changes in the August 1, 2023 update:
- New GPU for customers in China, Hong Kong and Macau:
- ThinkSystem NVIDIA H800 80GB PCIe Gen5 Passive GPU, 4X67A86451
First published: May 5, 2023
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.
