GPU Benchmarks NVIDIA A100 80 GB (SXM4) vs. NVIDIA H100 NVL (SXM5)
NVIDIA A100 80 GB (SXM4)
NVIDIA H100 NVL (SXM5)
Quick links:
- Benchmarks
- Specifications
- Deep Learning GPU Benchmarks 2024–2025
- 3D, GPU Rendering Benchmarks 2024–2025
Best GPUs for deep learning, AI development, compute in 2023–2024. Recommended GPU & hardware for AI training, inference (LLMs, generative AI). GPU training, inference benchmarks using PyTorch, TensorFlow for computer vision (CV), NLP, text-to-speech, etc.
We benchmark NVIDIA A100 80 GB (SXM4) vs NVIDIA H100 NVL (SXM5) GPUs and compare AI performance (deep learning training; FP16, FP32, PyTorch, TensorFlow), 3d rendering, Cryo-EM performance in the most popular apps (Octane, VRay, Redshift, Blender, Luxmark, Unreal Engine, Relion Cryo-EM).
Our benchmarks will help you decide which GPU (NVIDIA RTX 4090/4080, H100 Hopper, H200, A100, RTX 6000 Ada, A6000, A5000, or RTX 6000 ADA Lovelace) is the best GPU for your needs. We provide an in-depth analysis of the AI performance of each graphic card's performance so you can make the most informed decision possible. We offer deep learning and 3d rendering benchmarks that will help you get the most out of your hardware.
Looking for a GPU workstation or server for AI/ML, design, rendering, simulation or molecular dynamics?
Explore BIZON AI workstations or GPU servers.
Contact us today or explore our various customizable AI solutions.
Featured GPU benchmarks:
- NVIDIA H100 vs. A100 vs. RTX 6000 Ada
- NVIDIA RTX 4090 vs. RTX 4080 vs. RTX 3090
- NVIDIA RTX 3080 vs. 3080 Ti vs. RTX 3090
- NVIDIA RTX 3090 vs. NVIDIA A6000
- NVIDIA A6000 vs. A5000 vs. NVIDIA RTX 3090
- NVIDIA A100 vs. NVIDIA A6000
- NVIDIA RTX 2080 Ti vs. Titan RTX vs Quadro RTX8000
Benchmarks
Deep Learning GPU Benchmarks 2024–2025 [Updated]↑
As of May 2025
Resnet50 (FP16)
1 GPU
NVIDIA A100 80 GB (SXM4)
2606 points
NVIDIA H100 NVL (SXM5)
4197 points
4 GPU
NVIDIA A100 80 GB (SXM4)
9835 points
NVIDIA H100 NVL (SXM5)
15945 points
8 GPU
NVIDIA A100 80 GB (SXM4)
22552 points
NVIDIA H100 NVL (SXM5)
41496 points
higher is better
Resnet50 (FP32)
1 GPU
NVIDIA A100 80 GB (SXM4)
1189 points
NVIDIA H100 NVL (SXM5)
1903 points
4 GPU
NVIDIA A100 80 GB (SXM4)
4487 points
NVIDIA H100 NVL (SXM5)
7828 points
8 GPU
NVIDIA A100 80 GB (SXM4)
10240 points
NVIDIA H100 NVL (SXM5)
20616 points
higher is better
Resnet152 (FP16)
1 GPU
NVIDIA A100 80 GB (SXM4)
1056 points
NVIDIA H100 NVL (SXM5)
1701 points
4 GPU
NVIDIA A100 80 GB (SXM4)
3985 points
NVIDIA H100 NVL (SXM5)
6461 points
8 GPU
NVIDIA A100 80 GB (SXM4)
9308 points
NVIDIA H100 NVL (SXM5)
17128 points
higher is better
Resnet152 (FP32)
1 GPU
NVIDIA A100 80 GB (SXM4)
458 points
NVIDIA H100 NVL (SXM5)
733 points
4 GPU
NVIDIA A100 80 GB (SXM4)
1728 points
NVIDIA H100 NVL (SXM5)
3015 points
8 GPU
NVIDIA A100 80 GB (SXM4)
4138 points
NVIDIA H100 NVL (SXM5)
8331 points
higher is better
Inception V3 (FP16)
1 GPU
NVIDIA A100 80 GB (SXM4)
1572 points
NVIDIA H100 NVL (SXM5)
2532 points
4 GPU
NVIDIA A100 80 GB (SXM4)
5932 points
NVIDIA H100 NVL (SXM5)
9618 points
8 GPU
NVIDIA A100 80 GB (SXM4)
11871 points
NVIDIA H100 NVL (SXM5)
21842 points
higher is better
Inception V3 (FP32)
1 GPU
NVIDIA A100 80 GB (SXM4)
752 points
NVIDIA H100 NVL (SXM5)
1204 points
4 GPU
NVIDIA A100 80 GB (SXM4)
2839 points
NVIDIA H100 NVL (SXM5)
4952 points
8 GPU
NVIDIA A100 80 GB (SXM4)
6840 points
NVIDIA H100 NVL (SXM5)
13771 points
higher is better
Inception V4 (FP16)
1 GPU
NVIDIA A100 80 GB (SXM4)
727 points
NVIDIA H100 NVL (SXM5)
1170 points
4 GPU
NVIDIA A100 80 GB (SXM4)
2742 points
NVIDIA H100 NVL (SXM5)
4446 points
8 GPU
NVIDIA A100 80 GB (SXM4)
5115 points
NVIDIA H100 NVL (SXM5)
9411 points
higher is better
Inception V4 (FP32)
1 GPU
NVIDIA A100 80 GB (SXM4)
334 points
NVIDIA H100 NVL (SXM5)
535 points
4 GPU
NVIDIA A100 80 GB (SXM4)
1262 points
NVIDIA H100 NVL (SXM5)
2202 points
8 GPU
NVIDIA A100 80 GB (SXM4)
3126 points
NVIDIA H100 NVL (SXM5)
6294 points
higher is better
VGG16 (FP16)
1 GPU
NVIDIA A100 80 GB (SXM4)
930 points
NVIDIA H100 NVL (SXM5)
1497 points
4 GPU
NVIDIA A100 80 GB (SXM4)
3508 points
NVIDIA H100 NVL (SXM5)
5687 points
8 GPU
NVIDIA A100 80 GB (SXM4)
11931 points
NVIDIA H100 NVL (SXM5)
21953 points
higher is better
VGG16 (FP32)
1 GPU
NVIDIA A100 80 GB (SXM4)
684 points
NVIDIA H100 NVL (SXM5)
1095 points
4 GPU
NVIDIA A100 80 GB (SXM4)
2581 points
NVIDIA H100 NVL (SXM5)
4503 points
8 GPU
NVIDIA A100 80 GB (SXM4)
6093 points
NVIDIA H100 NVL (SXM5)
12267 points
higher is better
3D, GPU Rendering Benchmarks 2024–2025 [Updated]↑
As of May 2025
Blender
1 GPU
NVIDIA A100 80 GB (SXM4)
3841.95 score
NVIDIA H100 NVL (SXM5)
n/a
higher is better
Technical Specifications
Board Design
| NVIDIA A100 80 GB (SXM4) | NVIDIA H100 NVL (SXM5) | Difference | |
|---|---|---|---|
| Outputs | No outputs | No outputs | - |
| Power Connectors | None | None | - |
| Slot width | N/A | N/A | - |
| TDP | 400 W | 700 W | 300 W (75%) |
Clock Speeds
| NVIDIA A100 80 GB (SXM4) | NVIDIA H100 NVL (SXM5) | Difference | |
|---|---|---|---|
| Boost Clock | 1410 MHz | 1837 MHz | 427 MHz (30%) |
| GPU Clock | 1275 MHz | 1665 MHz | 390 MHz (31%) |
| Memory Clock | 3200 MHz | 5300 MHz | 2100 MHz (66%) |
Graphics Card
| NVIDIA A100 80 GB (SXM4) | NVIDIA H100 NVL (SXM5) | Difference | |
|---|---|---|---|
| Bus Interface | PCIe 4.0 x16 | PCIe 5.0 x16 | |
| Generation | Tesla (Axx) | Server Hopper (Hxx) |
Graphics Features
| NVIDIA A100 80 GB (SXM4) | NVIDIA H100 NVL (SXM5) | Difference | |
|---|---|---|---|
| OpenCL | 3 | 3 | - |
| CUDA | 8 | 9 |
Graphics Processor
| NVIDIA A100 80 GB (SXM4) | NVIDIA H100 NVL (SXM5) | Difference | |
|---|---|---|---|
| Architecture | Ampere | Hopper | |
| Die Size | 826 mm2 | 814 mm2 | -12 mm2 (-1%) |
| GPU Name | GA100 | GH100 | |
| Process Size | 7 nm | 5 nm | -2 nm (-29%) |
| Transistors | 54200 million | 80000 million | 25800 million (48%) |
Memory
| NVIDIA A100 80 GB (SXM4) | NVIDIA H100 NVL (SXM5) | Difference | |
|---|---|---|---|
| Bandwidth | 2039 GB/s | 3360 GB/s | 1321 GB/s (65%) |
| Memory Bus | 5120 bit | 5120 bit | - |
| Memory Size | 80 GB | 96 GB | 16 GB (20%) |
| Memory Type | HBM2e | HBM3 |
Render Config
| NVIDIA A100 80 GB (SXM4) | NVIDIA H100 NVL (SXM5) | Difference | |
|---|---|---|---|
| ROPs | 160 | 24 | -136 (-85%) |
| Shading Units/ CUDA Cores | 6912 | 16896 | 9984 (144%) |
| TMUs | 432 | 528 | 96 (22%) |
| Tensor Cores | 432 | 528 | 96 (22%) |
Theoretical Performance
| NVIDIA A100 80 GB (SXM4) | NVIDIA H100 NVL (SXM5) | Difference | |
|---|---|---|---|
| FP16 (half) performance | 77.97 TFLOPS | 248.3 TFLOPS | 170.33 TFLOPS (218%) |
| FP32 (float) performance | 19.49 TFLOPS | 62.08 TFLOPS | 42.59 TFLOPS (219%) |
| FP64 (double) performance | 9746 GFLOPS | 31040 GFLOPS | 21294 GFLOPS (218%) |
| Pixel Rate | 225.6 GPixel/s | 44.09 GPixel/s | -181.51 GPixel/s (-80%) |
| Texture Rate | 609.1 GTexel/s | 969.9 GTexel/s | 360.8 GTexel/s (59%) |
Price
| NVIDIA A100 80 GB (SXM4) | NVIDIA H100 NVL (SXM5) | Difference | |
|---|---|---|---|
| Release Date | Nov 16th, 2020 | Mar 21st, 2023 |
Test bench configuration
| NVIDIA A100 80 GB (SXM4) | NVIDIA H100 NVL (SXM5) | Difference | |
|---|---|---|---|
| Hardware | BIZON X9000 More details | BIZON X5000 More details | |
| Software | 3D Rendering: Nvidia Driver: 461.09 VRay Benchmark: 5 Octane Benchmark: 2020.1.5 Redshift Benchmark: 3.0.28 Demo Blender: 2.90 Luxmark: 3.1 | 3D Rendering: VRay Benchmark: 5 Octane Benchmark: 2020.1.5 Redshift Benchmark: 3.0.28 Demo Blender: 2.90 Luxmark: 3.1 |
Recommended hardware
| NVIDIA A100 80 GB (SXM4) | NVIDIA H100 NVL (SXM5) | Difference | |
|---|---|---|---|
| Best NVIDIA GPU servers | BIZON X9000 | - |
