GEFORCE

NVIDIA A100 SXM4 80 GB

NVIDIA graphics card specifications and benchmark scores

80 GB
VRAM
1410
MHz Boost
400W
TDP
5120
Bus Width
Tensor Cores

At a Glance

NVIDIA
VRAM 80 GB
Boost Clock 1,410 MHz
Shaders 6,912
Bus Width 5120-bit
TDP 400W
Memory Type HBM2e
Architecture Ampere
nm
Process 7 nm
Released Nov 2020

NVIDIA A100 SXM4 80 GB Specifications

A100 SXM4 80 GB GPU Core

Shader units and compute resources

The NVIDIA A100 SXM4 80 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.

Shading Units
6,912
Shaders
6,912
TMUs
432
ROPs
160
SM Count
108

A100 SXM4 80 GB Clock Speeds

GPU and memory frequencies

Clock speeds directly impact the A100 SXM4 80 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 A100 SXM4 80 GB by NVIDIA dynamically adjusts frequencies based on workload, temperature, and power limits to maximize performance while maintaining stability.

Base Clock
1275 MHz
Base Clock
1,275 MHz
Boost Clock
1410 MHz
Boost Clock
1,410 MHz
Memory Clock
1593 MHz 3.2 Gbps effective
GDDR GDDR 6X 6X

NVIDIA's A100 SXM4 80 GB Memory

VRAM capacity and bandwidth

VRAM (Video RAM) is dedicated memory for storing textures, frame buffers, and shader data. The A100 SXM4 80 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.

Memory Size
80 GB
VRAM
81,920 MB
Memory Type
HBM2e
VRAM Type
HBM2e
Memory Bus
5120 bit
Bus Width
5120-bit
Bandwidth
2.04 TB/s

A100 SXM4 80 GB by NVIDIA Cache

On-chip cache hierarchy

On-chip cache provides ultra-fast data access for the A100 SXM4 80 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.

L1 Cache
192 KB (per SM)
L2 Cache
40 MB

A100 SXM4 80 GB Theoretical Performance

Compute and fill rates

Theoretical performance metrics provide a baseline for comparing the NVIDIA A100 SXM4 80 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.

FP32 (Float)
19.49 TFLOPS
FP64 (Double)
9.746 TFLOPS (1:2)
FP16 (Half)
77.97 TFLOPS (4:1)
Pixel Rate
225.6 GPixel/s
Texture Rate
609.1 GTexel/s

A100 SXM4 80 GB Ray Tracing & AI

Hardware acceleration features

The NVIDIA A100 SXM4 80 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 A100 SXM4 80 GB capable of delivering both stunning graphics and smooth frame rates in modern titles.

Tensor Cores
432
BF16
311.84 TFLOPS (16:1)
TF32
155.92 TFLOPs (8:1)

Ampere Architecture & Process

Manufacturing and design details

The NVIDIA A100 SXM4 80 GB is built on NVIDIA's Ampere 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 A100 SXM4 80 GB will perform in GPU benchmarks compared to previous generations.

Architecture
Ampere
GPU Name
GA100
Process Node
7 nm
Foundry
TSMC
Transistors
54,200 million
Die Size
826 mm²
Density
65.6M / mm²

NVIDIA's A100 SXM4 80 GB Power & Thermal

TDP and power requirements

Power specifications for the NVIDIA A100 SXM4 80 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 A100 SXM4 80 GB to maintain boost clocks without throttling.

TDP
400 W
TDP
400W
Power Connectors
None
Suggested PSU
800 W

A100 SXM4 80 GB by NVIDIA Physical & Connectivity

Dimensions and outputs

Physical dimensions of the NVIDIA A100 SXM4 80 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.

Slot Width
OAM Module
Bus Interface
PCIe 4.0 x16
Display Outputs
No outputs
Display Outputs
No outputs

NVIDIA API Support

Graphics and compute APIs

API support determines which games and applications can fully utilize the NVIDIA A100 SXM4 80 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.

OpenCL
3.0
CUDA
8.0

A100 SXM4 80 GB Product Information

Release and pricing details

The NVIDIA A100 SXM4 80 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 A100 SXM4 80 GB by NVIDIA represents good value at current market prices. Predecessor and successor information aids in tracking generational improvements and planning future upgrades.

Manufacturer
NVIDIA
Release Date
Nov 2020
Production
End-of-life
Predecessor
Tesla Turing
Successor
Server Ada

A100 SXM4 80 GB Benchmark Scores

geekbench_vulkanSource

Geekbench Vulkan tests GPU compute using the modern low-overhead Vulkan API. This shows how NVIDIA A100 SXM4 80 GB performs with next-generation graphics and compute workloads. Vulkan offers better CPU efficiency than older APIs like OpenGL.

geekbench_vulkan #21 of 444
183,725
49%
Max: 376,915

About NVIDIA A100 SXM4 80 GB

The NVIDIA A100 SXM4 80 GB is an end-of-life server accelerator built on the Ampere architecture, using the GA100 chip fabricated on a 7 nm process at TSMC. It contains 54,200 million transistors on an 826 mm² die, yielding a transistor density of 65.6M per mm². In the benchmark database, it holds a 50th percentile ranking against all GPUs, with an average benchmark score of 0, indicating that its profile is defined by raw specifications rather than measured performance data.

Memory Subsystem

The A100 SXM4 80 GB pairs 80 GB of HBM2e memory with a 5120-bit bus, delivering a peak bandwidth of 2.04 TB/s. The memory clock runs at 1593 MHz, which translates to 3.2 Gbps effective. This configuration provides an exceptionally wide path for data movement, and the sheer capacity allows large datasets to reside on-chip. For high-resolution workloads, the 80 GB pool reduces reliance on system memory, while the 2.04 TB/s bandwidth ensures that the shading units and tensor cores are fed without bottlenecks. The 5120-bit bus is among the widest in the database, and the bandwidth figure is a direct indicator of its compute-class positioning. In practice, applications demanding large frame buffers or massive model weights can operate efficiently, though the absence of display outputs means this capability is aimed at compute tasks rather than direct rendering output.

Ray Tracing and Feature Set

The fact pack does not specify a dedicated ray tracing core count for this accelerator. Instead, the compute feature set is anchored by 432 tensor cores, which are designed for matrix operations and AI workloads. The FP16 throughput is listed at 77.97 TFLOPS with a 4:1 ratio, while FP32 reaches 19.49 TFLOPS. This asymmetry indicates a strong bias toward mixed-precision and tensor-heavy tasks. The API support fields for DirectX, OpenGL, and Vulkan are all listed as null, meaning the database records no API compatibility for this part. Additionally, the card has no display outputs, reinforcing its role as a compute-only server module. The pixel rate is 225.6 GPixel/s and the texture rate is 609.1 GTexel/s, which are rasterization metrics, but without RT cores or display outputs, the feature set is clearly optimized for data center compute rather than real-time graphics.

Benchmark Performance

The database records an average benchmark score of 0 for the NVIDIA A100 SXM4 80 GB, and its percentile ranking against all GPUs is 50. This places it exactly at the median of the entire GPU distribution in the database. Because no benchmark scores are populated, the performance analysis must rely on the raw compute metrics. The FP32 throughput of 19.49 TFLOPS represents the single-precision peak, while the FP16 peak of 77.97 TFLOPS (4:1) indicates a significant acceleration for half-precision workloads. The texture rate of 609.1 GTexel/s and pixel rate of 225.6 GPixel/s provide additional context for fill-rate-bound scenarios. However, without direct rival scores or delta percentages, it is not possible to quantify its standing against specific competitors. The 50th percentile suggests a mid-pack position in the overall database, but the absence of benchmark data means this ranking is derived from the aggregated hardware profile rather than measured tests. The 432 tensor cores are the most prominent compute resource, and their presence suggests that the intended performance envelope is in the AI and scientific computing domain.

Who Should Consider It

Given the server-oriented design, the A100 SXM4 80 GB is suited for environments where memory capacity and bandwidth are paramount. The 80 GB HBM2e pool and 2.04 TB/s bandwidth make it a candidate for high-resolution rendering tasks that require large texture sets or for training large neural networks that demand extensive parameter storage. The 400 W TDP and OAM module form factor indicate that it is designed for dense server installations, not desktop systems. Users with workloads that can leverage the FP16 4:1 ratio will find the 77.97 TFLOPS peak useful for mixed-precision training. Conversely, those needing display output or real-time ray tracing will find no support, as the part has no display outputs and no RT core count is specified. The end-of-life production status suggests that new deployments should consider the successor, Server Ada, but the A100 remains a viable option for existing infrastructure. The absence of benchmark scores means that purchasing decisions should be based on the raw compute specifications and the known compatibility with server platforms.

How It Compares

The database lists no nearest rivals for the NVIDIA A100 SXM4 80 GB. Consequently, there are no direct comparison scores or delta percentages to analyze. Its predecessor is the Tesla Turing generation, and its successor is Server Ada. Without rival data, the comparison must rely on architectural lineage. The A100 is a member of the Server Ampere (Axx) generation, which follows Tesla Turing. The transition from Turing to Ampere brought the GA100 chip and the 7 nm process, but the database does not provide specific performance deltas. Similarly, the successor Server Ada is noted, but no metrics are available to quantify the generational leap. In the absence of rival scores, the A100's position is defined by its own specifications: the 5120-bit memory bus, 432 tensor cores, and 19.49 TFLOPS FP32 peak. These figures place it in the high-compute tier, but the lack of benchmark data prevents a precise ranking against other accelerators.

Power and Cooling

The NVIDIA A100 SXM4 80 GB has a thermal design power of 400 W. The suggested power supply rating is 800 W, which provides a comfortable margin for the system. The power connector field is listed as "None," which is consistent with the OAM Module slot width. This indicates that power is delivered through the module interface rather than through standard PCIe power connectors. The absence of a dedicated connector simplifies installation in compatible server chassis. The 400 W TDP requires adequate cooling, but the fact pack does not specify a cooler type. Given the OAM form factor, cooling is typically provided by the server chassis, either through passive heatsinks with active airflow or integrated liquid cooling loops. The 800 W PSU recommendation is a system-level requirement, meaning the entire server platform should be sized to accommodate this draw. Since no display outputs are present, the card is entirely compute-focused, and power delivery must be stable to support sustained FP16 or FP32 workloads.

FAQ

Q: What is the memory bandwidth of the NVIDIA A100 SXM4 80 GB?

A: The memory bandwidth is 2.04 TB/s, provided by 80 GB of HBM2e memory on a 5120-bit bus.

Q: Does the A100 SXM4 80 GB have any display outputs?

A: No, the fact pack lists "No outputs" for display outputs, confirming it is a compute-only server module.

Q: What is the thermal design power (TDP) and suggested power supply?

A: The TDP is 400 W, and the suggested PSU rating is 800 W.

Q: What is the process node and foundry for this chip?

A: It is fabricated on a 7 nm process at TSMC.

Q: How many tensor cores does it have?

A: It has 432 tensor cores.

Q: What is the release date and production status?

A: It was released on November 15, 2020, and its production status is listed as end-of-life.

The AMD Equivalent of A100 SXM4 80 GB

Looking for a similar graphics card from AMD? The AMD Radeon RX 6800 XT offers comparable performance and features in the AMD lineup.

AMD Radeon RX 6800 XT

AMD • 16 GB VRAM

View Specs Compare

Popular NVIDIA A100 SXM4 80 GB Comparisons

See how the A100 SXM4 80 GB stacks up against similar graphics cards from the same generation and competing brands.

Compare A100 SXM4 80 GB with Other GPUs

Select another GPU to compare specifications and benchmarks side-by-side.

Browse GPUs