NVIDIA H100 SXM5 94 GB
NVIDIA graphics card specifications and benchmark scores
Tensor CoresAt a Glance
NVIDIANVIDIA H100 SXM5 94 GB Specifications
H100 SXM5 94 GB GPU Core
Shader units and compute resources
The NVIDIA H100 SXM5 94 GB GPU core specifications define its raw processing power for graphics and compute workloads. Shading units (also called CUDA cores, stream processors, or execution units depending on manufacturer) handle the parallel calculations required for rendering. TMUs (Texture Mapping Units) process texture data, while ROPs (Render Output Units) handle final pixel output. Higher shader counts generally translate to better GPU benchmark performance, especially in demanding games and 3D applications.
H100 SXM5 94 GB Clock Speeds
GPU and memory frequencies
Clock speeds directly impact the H100 SXM5 94 GB's performance in GPU benchmarks and real-world gaming. The base clock represents the minimum guaranteed frequency, while the boost clock indicates peak performance under optimal thermal conditions. Memory clock speed affects texture loading and frame buffer operations. The H100 SXM5 94 GB by NVIDIA dynamically adjusts frequencies based on workload, temperature, and power limits to maximize performance while maintaining stability.
NVIDIA's H100 SXM5 94 GB Memory
VRAM capacity and bandwidth
VRAM (Video RAM) is dedicated memory for storing textures, frame buffers, and shader data. The H100 SXM5 94 GB's memory capacity determines how well it handles high-resolution textures and multiple displays. Memory bandwidth, measured in GB/s, affects how quickly data moves between the GPU and VRAM. Higher bandwidth improves performance in memory-intensive scenarios like 4K gaming. The memory bus width and type (GDDR6, GDDR6X, HBM) significantly influence overall GPU benchmark scores.
H100 SXM5 94 GB by NVIDIA Cache
On-chip cache hierarchy
On-chip cache provides ultra-fast data access for the H100 SXM5 94 GB, reducing the need to fetch data from slower VRAM. L1 and L2 caches store frequently accessed data close to the compute units. AMD's Infinity Cache (L3) dramatically increases effective bandwidth, improving GPU benchmark performance without requiring wider memory buses. Larger cache sizes help maintain high frame rates in memory-bound scenarios and reduce power consumption by minimizing VRAM accesses.
H100 SXM5 94 GB Theoretical Performance
Compute and fill rates
Theoretical performance metrics provide a baseline for comparing the NVIDIA H100 SXM5 94 GB against other graphics cards. FP32 (single-precision) performance, measured in TFLOPS, indicates compute capability for gaming and general GPU workloads. FP64 (double-precision) matters for scientific computing. Pixel and texture fill rates determine how quickly the GPU can render complex scenes. While real-world GPU benchmark results depend on many factors, these specifications help predict relative performance levels.
H100 SXM5 94 GB Ray Tracing & AI
Hardware acceleration features
The NVIDIA H100 SXM5 94 GB includes dedicated hardware for ray tracing and AI acceleration. RT cores handle real-time ray tracing calculations for realistic lighting, reflections, and shadows in supported games. Tensor cores (NVIDIA) or XMX cores (Intel) accelerate AI workloads including DLSS, FSR, and XeSS upscaling technologies. These features enable higher visual quality without proportional performance costs, making the H100 SXM5 94 GB capable of delivering both stunning graphics and smooth frame rates in modern titles.
Hopper Architecture & Process
Manufacturing and design details
The NVIDIA H100 SXM5 94 GB is built on NVIDIA's Hopper architecture, which defines how the GPU processes graphics and compute workloads. The manufacturing process node affects power efficiency, thermal characteristics, and maximum clock speeds. Smaller process nodes pack more transistors into the same die area, enabling higher performance per watt. Understanding the architecture helps predict how the H100 SXM5 94 GB will perform in GPU benchmarks compared to previous generations.
NVIDIA's H100 SXM5 94 GB Power & Thermal
TDP and power requirements
Power specifications for the NVIDIA H100 SXM5 94 GB determine PSU requirements and thermal management needs. TDP (Thermal Design Power) indicates the heat output under typical loads, guiding cooler selection. Power connector requirements ensure adequate power delivery for stable operation during demanding GPU benchmarks. The suggested PSU wattage accounts for the entire system, not just the graphics card. Efficient power delivery enables the H100 SXM5 94 GB to maintain boost clocks without throttling.
H100 SXM5 94 GB by NVIDIA Physical & Connectivity
Dimensions and outputs
Physical dimensions of the NVIDIA H100 SXM5 94 GB are critical for case compatibility. Card length, height, and slot width determine whether it fits in your chassis. The PCIe interface version affects bandwidth for communication with the CPU. Display outputs define monitor connectivity options, with modern cards supporting multiple high-resolution displays simultaneously. Verify these specifications against your case and motherboard before purchasing to ensure a proper fit.
NVIDIA API Support
Graphics and compute APIs
API support determines which games and applications can fully utilize the NVIDIA H100 SXM5 94 GB. DirectX 12 Ultimate enables advanced features like ray tracing and variable rate shading. Vulkan provides cross-platform graphics capabilities with low-level hardware access. OpenGL remains important for professional applications and older games. CUDA (NVIDIA) and OpenCL enable GPU compute for video editing, 3D rendering, and scientific applications. Higher API versions unlock newer graphical features in GPU benchmarks and games.
H100 SXM5 94 GB Product Information
Release and pricing details
The NVIDIA H100 SXM5 94 GB is manufactured by NVIDIA as part of their graphics card lineup. Release date and launch pricing provide context for comparing GPU benchmark results with competing products from the same era. Understanding the product lifecycle helps evaluate whether the H100 SXM5 94 GB by NVIDIA represents good value at current market prices. Predecessor and successor information aids in tracking generational improvements and planning future upgrades.
H100 SXM5 94 GB Benchmark Scores
No benchmark data available for this GPU.
About NVIDIA H100 SXM5 94 GB
The NVIDIA H100 SXM5 94 GB is a Server Hopper (Hxx) entry built around the GH100 chip from NVIDIA, manufactured by TSMC on a 5 nm process. The data lists 80,000 million transistors on an 814 mm² die, for a transistor density of 98.3M / mm². The memory subsystem is 94 GB HBM3 on a 5120-bit bus with 3.36 TB/s bandwidth and a memory clock of 1313 MHz / 5.3 Gbps effective. Compute resources are 16896 shading units, 528 TMUs, 24 ROPs, and 528 tensor cores, with peak FP32 throughput of 66.91 TFLOPS and peak FP16 throughput of 267.6 TFLOPS (4:1). The module is rated for a 700 W TDP, carries an 8-pin EPS power connector, has a suggested PSU of 1100 W, uses an SXM Module slot width, connects through PCIe 5.0 x16, and has no display outputs. The release timestamp is 2023-03-20T17:00:00.000Z; the benchmarks array is empty, avgBenchmarkScore is 0, percentileVsAllGpus is 50, and nearestRivals is empty.
Benchmark Performance
Because the benchmarks array is empty and nearestRivals is empty, there are no measured scores and no deltaPct values to analyze. The only numeric performance-position fields are avgBenchmarkScore: 0 and percentileVsAllGpus: 50. A zero average score with no benchmark entries means the 50 value cannot be tied to an actual computed workload; it appears in the record as a stored figure rather than as a result of the listed benchmarks.
Without rival percentages, the quantitative performance description must come from peak throughput fields. The FP32 rate is 66.91 TFLOPS. The FP16 rate is 267.6 TFLOPS (4:1), which in the data is expressed as a 4:1 ratio relative to the FP32 path. The texture rate is 1,045.4 GTexel/s and the pixel rate is 47.52 GPixel/s, alongside base and boost clocks of 1350 MHz and 1980 MHz. Memory performance is 94 GB HBM3 on a 5120-bit bus with 3.36 TB/s bandwidth and a memory clock of 1313 MHz / 5.3 Gbps effective. That bandwidth figure is the memory-side throughput that would influence large-data workloads, but no application-level score is attached to it.
The record contains no benchmark result, so no application-level percentage ahead of or behind a named rival can be stated. The percentileVsAllGpus of 50 is the only ranking-like field, but with an avgBenchmarkScore of 0 in the same record, the database position is not supported by a score. Therefore, exact percentage deltas against nearest rivals cannot be computed; the data includes zero rivals and zero benchmark entries.
Who Should Consider It
No display outputs are listed, so the data does not describe a product for direct monitor connection. The slot width is SXM Module, and the host interface is PCIe 5.0 x16; those two fields position it as a module for a server-class system. The database contains no resolution-specific or display-oriented frame-rate scores, so there are no measured FPS values from which to recommend a resolution or settings tier.
The 16896 shading units, 528 TMUs, 24 ROPs, 94 GB HBM3, and 3.36 TB/s bandwidth are the resources on record, but they do not by themselves produce a frame-rate conclusion. The 528 tensor cores and 267.6 TFLOPS FP16 (4:1) point toward tensor-heavy workloads, though no tensor benchmark score is present. For a user choosing by measured outcomes, this record offers no such outcome; for a user choosing by capacity and peak rate, the memory capacity and bandwidth are the standout values.
Because no benchmark scores accompany the 50 percentile field, a resolution-specific buyer cannot be mapped to a quality level. The absence of API fields — DirectX, OpenGL, and Vulkan are all null — further prevents a traditional display-API compatibility assessment. The specification suggests a compute-oriented module without attached display outputs; it does not provide evidence for a display-attached use case at any settings level.
Power and Cooling
The thermal and power figures in the record are a 700 W TDP, a suggested PSU of 1100 W, and an 8-pin EPS power connector. The slot width field is SXM Module, which is the physical form factor represented in the data; the power connector field does not list any other connector type. The memory clock of 1313 MHz / 5.3 Gbps effective and HBM3 memory type are part of the memory subsystem that operates within that TDP. The manufacturing context is TSMC's 5 nm process with 80,000 million transistors and an 814 mm² die.
No cooling solution size or thermal solution is listed in the data, so the only numerical anchor for cooling is the 700 W TDP. The 1100 W suggested PSU is the supply-side recommendation; users of an SXM Module would rely on the system's power delivery rather than a standalone card's auxiliary connectors. The data does not provide a second power connector or a separate maximum power figure, so the 700 W TDP and 8-pin EPS connector stand as the complete power picture.
How It Compares
The nearestRivals array in the data is empty, so there is no named competitor with a score or deltaPct to compare. This means no rival-by-rival paragraph can be built from measured values. The only relative fields are percentileVsAllGpus: 50 and avgBenchmarkScore: 0; since there are no benchmark entries, these two fields do not form a measurable ranking against another product.
The predecessor label is Server Ada and the successor label is Server Blackwell, but the data stores no score for either, so no generational percentage can be calculated. If the empty nearestRivals list is taken at face value, the database does not currently place this part next to any nearest rival. Consequently, the comparison table that would normally list product names, scores, and deltaPct values is absent, and the position of the H100 SXM5 94 GB relative to other GPUs must remain unstated in this record.
Ray Tracing and Feature Set
The data lists rtCores as null, so no ray-tracing core count is provided. In contrast, the tensor core count is 528; the FP16 peak rate of 267.6 TFLOPS (4:1) accompanies those tensor cores in the specification. The GPU is the GH100 chip in NVIDIA's Hopper architecture, with generation listed as Server Hopper (Hxx). API support is not quantified: DirectX, OpenGL, and Vulkan are all null. The display output field is "No outputs," which is consistent with a module that does not expose a video path.
Other feature-level facts include 94 GB HBM3 with a 5120-bit bus and 3.36 TB/s bandwidth, PCIe 5.0 x16 host interface, and 5 nm TSMC manufacturing with 80,000 million transistors and an 814 mm² die. Because the rtCores field is null and the API fields are null, the data does not authorize any specific ray-tracing performance or compatibility statement. The only dedicated acceleration feature with a count is the tensor core block: 528 tensor cores. Texture and pixel rates are 1,045.4 GTexel/s and 47.52 GPixel/s, respectively, but these are rate values, not ray-tracing scores.
FAQ
Q: What memory does the NVIDIA H100 SXM5 94 GB use?
A: It uses 94 GB of HBM3 on a 5120-bit bus with a memory clock of 1313 MHz / 5.3 Gbps effective and a bandwidth of 3.36 TB/s.
Q: What is the TDP and PSU recommendation in the data?
A: The TDP is 700 W, the suggested PSU is 1100 W, and the power connector is 8-pin EPS.
Q: Does this module provide display outputs?
A: No. The displayOutputs field is "No outputs," and the slot width is SXM Module.
Q: How many tensor cores and RT cores are listed?
A: Tensor cores are listed as 528; rtCores is null in the data, so no RT core count is given.
Q: What are the listed core counts and peak rates?
A: The record shows 16896 shading units, 528 TMUs, 24 ROPs, 66.91 TFLOPS FP32, 267.6 TFLOPS FP16 (4:1), 1,045.4 GTexel/s texture rate, and 47.52 GPixel/s pixel rate.
Q: When was it released and what is its generation?
A: The release timestamp is 2023-03-20T17:00:00.000Z, generation is Server Hopper (Hxx), predecessor is Server Ada, and successor is Server Blackwell.
The AMD Equivalent of H100 SXM5 94 GB
Looking for a similar graphics card from AMD? The AMD Radeon RX 7600 offers comparable performance and features in the AMD lineup.
Popular NVIDIA H100 SXM5 94 GB Comparisons
See how the H100 SXM5 94 GB stacks up against similar graphics cards from the same generation and competing brands.
Compare H100 SXM5 94 GB with Other GPUs
Select another GPU to compare specifications and benchmarks side-by-side.
Browse GPUs