GPU Benchmarks NVIDIA A100 40 GB (PCIe) vs. NVIDIA RTX 4090
NVIDIA A100 40 GB (PCIe)
NVIDIA RTX 4090
Quick links:
- Benchmarks
- Specifications
- Deep Learning GPU Benchmarks 2024–2025
- 3D, GPU Rendering Benchmarks 2024–2025
- RELION Cryo-EM Benchmarks 2024-2025
- Llama3 70B Inference Benchmark 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 40 GB (PCIe) vs NVIDIA RTX 4090 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 40 GB (PCIe)
2179 points
NVIDIA RTX 4090
1720 points
4 GPU
NVIDIA A100 40 GB (PCIe)
8561 points
NVIDIA RTX 4090
5934 points
8 GPU
NVIDIA A100 40 GB (PCIe)
16797 points
NVIDIA RTX 4090
n/a
higher is better
Resnet50 (FP32)
1 GPU
NVIDIA A100 40 GB (PCIe)
1001 points
NVIDIA RTX 4090
927 points
4 GPU
NVIDIA A100 40 GB (PCIe)
3849 points
NVIDIA RTX 4090
1715 points
8 GPU
NVIDIA A100 40 GB (PCIe)
7557 points
NVIDIA RTX 4090
n/a
higher is better
Resnet152 (FP16)
1 GPU
NVIDIA A100 40 GB (PCIe)
930 points
NVIDIA RTX 4090
n/a
4 GPU
NVIDIA A100 40 GB (PCIe)
3557 points
NVIDIA RTX 4090
n/a
8 GPU
NVIDIA A100 40 GB (PCIe)
6809 points
NVIDIA RTX 4090
n/a
higher is better
Resnet152 (FP32)
1 GPU
NVIDIA A100 40 GB (PCIe)
409 points
NVIDIA RTX 4090
n/a
4 GPU
NVIDIA A100 40 GB (PCIe)
1498 points
NVIDIA RTX 4090
n/a
8 GPU
NVIDIA A100 40 GB (PCIe)
2851 points
NVIDIA RTX 4090
n/a
higher is better
Inception V3 (FP16)
1 GPU
NVIDIA A100 40 GB (PCIe)
1283 points
NVIDIA RTX 4090
n/a
4 GPU
NVIDIA A100 40 GB (PCIe)
5218 points
NVIDIA RTX 4090
n/a
8 GPU
NVIDIA A100 40 GB (PCIe)
10122 points
NVIDIA RTX 4090
n/a
higher is better
Inception V3 (FP32)
1 GPU
NVIDIA A100 40 GB (PCIe)
658 points
NVIDIA RTX 4090
n/a
4 GPU
NVIDIA A100 40 GB (PCIe)
2568 points
NVIDIA RTX 4090
n/a
8 GPU
NVIDIA A100 40 GB (PCIe)
5058 points
NVIDIA RTX 4090
n/a
higher is better
Inception V4 (FP16)
1 GPU
NVIDIA A100 40 GB (PCIe)
616 points
NVIDIA RTX 4090
n/a
4 GPU
NVIDIA A100 40 GB (PCIe)
2377 points
NVIDIA RTX 4090
n/a
8 GPU
NVIDIA A100 40 GB (PCIe)
4532 points
NVIDIA RTX 4090
n/a
higher is better
Inception V4 (FP32)
1 GPU
NVIDIA A100 40 GB (PCIe)
290 points
NVIDIA RTX 4090
n/a
4 GPU
NVIDIA A100 40 GB (PCIe)
1031 points
NVIDIA RTX 4090
n/a
8 GPU
NVIDIA A100 40 GB (PCIe)
1950 points
NVIDIA RTX 4090
n/a
higher is better
VGG16 (FP16)
1 GPU
NVIDIA A100 40 GB (PCIe)
1249 points
NVIDIA RTX 4090
n/a
4 GPU
NVIDIA A100 40 GB (PCIe)
4989 points
NVIDIA RTX 4090
n/a
8 GPU
NVIDIA A100 40 GB (PCIe)
10733 points
NVIDIA RTX 4090
n/a
higher is better
VGG16 (FP32)
1 GPU
NVIDIA A100 40 GB (PCIe)
529 points
NVIDIA RTX 4090
n/a
4 GPU
NVIDIA A100 40 GB (PCIe)
2215 points
NVIDIA RTX 4090
n/a
8 GPU
NVIDIA A100 40 GB (PCIe)
4278 points
NVIDIA RTX 4090
n/a
higher is better
3D, GPU Rendering Benchmarks 2024–2025 [Updated]↑
As of May 2025
V-Ray
1 GPU
NVIDIA A100 40 GB (PCIe)
1555 points
NVIDIA RTX 4090
5556 points
higher is better
Octane
1 GPU
NVIDIA A100 40 GB (PCIe)
498 points
NVIDIA RTX 4090
1445 points
higher is better
Redshift
1 GPU
NVIDIA A100 40 GB (PCIe)
n/a
NVIDIA RTX 4090
1.16 minutes
lower is better
Blender
1 GPU
NVIDIA A100 40 GB (PCIe)
3788 score
NVIDIA RTX 4090
12123.96 score
higher is better
Luxmark
1 GPU
NVIDIA A100 40 GB (PCIe)
n/a
NVIDIA RTX 4090
158815 points
higher is better
Unreal Engine
1 GPU
NVIDIA A100 40 GB (PCIe)
n/a
NVIDIA RTX 4090
92.1 FPS
higher is better
RELION Cryo-EM Benchmarks 2024-2025 [Updated]↑
As of May 2025
Total run time
1 GPU
NVIDIA A100 40 GB (PCIe)
178.9 Min
NVIDIA RTX 4090
105.2 Min
4 GPU
NVIDIA A100 40 GB (PCIe)
50.3 Min
NVIDIA RTX 4090
53.2 Min
lower is better
Llama3 70B Inference Benchmark 2024–2025 [Updated]↑
As of May 2025
Eval rate
1 GPU
NVIDIA A100 40 GB (PCIe)
n/a
NVIDIA RTX 4090
9.95 tokens/s
2 GPU
NVIDIA A100 40 GB (PCIe)
n/a
NVIDIA RTX 4090
19.99 tokens/s
higher is better
Technical Specifications
Board Design
| NVIDIA A100 40 GB (PCIe) | NVIDIA RTX 4090 | Difference | |
|---|---|---|---|
| Length | 11 in / 267 mm | 13 in / 336 mm | 3 in / 69 mm (26%) |
| Outputs | No outputs | 1x HDMI, 3x DisplayPort | |
| Power Connectors | 8-pin EPS | 1x 16-pin | |
| Slot width | Dual-slot | Triple-slot | |
| TDP | 250 W | 450 W | 200 W (80%) |
Clock Speeds
| NVIDIA A100 40 GB (PCIe) | NVIDIA RTX 4090 | Difference | |
|---|---|---|---|
| Boost Clock | 1410 MHz | 2520 MHz | 1110 MHz (79%) |
| GPU Clock | 765 MHz | 2235 MHz | 1470 MHz (192%) |
| Memory Clock | 2400 MHz | 21200 MHz | 18800 MHz (783%) |
Graphics Card
| NVIDIA A100 40 GB (PCIe) | NVIDIA RTX 4090 | Difference | |
|---|---|---|---|
| Bus Interface | PCIe 4.0 x16 | PCIe 4.0 x16 | - |
| Generation | Tesla (Axx) | GeForce 40 |
Graphics Features
| NVIDIA A100 40 GB (PCIe) | NVIDIA RTX 4090 | Difference | |
|---|---|---|---|
| OpenCL | 2 | 3 | |
| CUDA | 8 | 8.9 | |
| DirectX | - | 12 Ultimate (12_2) | |
| OpenGL | - | 4.6 | |
| Shader Model | - | 6.7 |
Graphics Processor
| NVIDIA A100 40 GB (PCIe) | NVIDIA RTX 4090 | Difference | |
|---|---|---|---|
| Architecture | Ampere | Ada Lovelace | |
| Die Size | 826 mm2 | 608 mm2 | -218 mm2 (-26%) |
| GPU Name | GA100 | AD102-300-A1 | |
| Process Size | 7 nm | 5 nm | -2 nm (-29%) |
| Transistors | 54200 million | 76300 million | 22100 million (41%) |
Memory
| NVIDIA A100 40 GB (PCIe) | NVIDIA RTX 4090 | Difference | |
|---|---|---|---|
| Bandwidth | 1555 GB/s | 1018 GB/s | -537 GB/s (-35%) |
| Memory Bus | 5120 bit | 384 bit | -4736 bit (-92%) |
| Memory Size | 40 GB | 24 GB | -16 GB (-40%) |
| Memory Type | HBM2e | GDDR6X |
Render Config
| NVIDIA A100 40 GB (PCIe) | NVIDIA RTX 4090 | Difference | |
|---|---|---|---|
| ROPs | 160 | 192 | 32 (20%) |
| Shading Units/ CUDA Cores | 6912 | 16384 | 9472 (137%) |
| TMUs | 432 | 512 | 80 (19%) |
| Tensor Cores | 432 | 512 | 80 (19%) |
| RT Cores | - | 128 |
Theoretical Performance
| NVIDIA A100 40 GB (PCIe) | NVIDIA RTX 4090 | Difference | |
|---|---|---|---|
| FP16 (half) performance | 77.97 TFLOPS | 82.58 TFLOPS | 4.61 TFLOPS (6%) |
| FP32 (float) performance | 19.49 TFLOPS | 82.58 TFLOPS | 63.09 TFLOPS (324%) |
| FP64 (double) performance | 9746 GFLOPS | 1290 GFLOPS | -8456 GFLOPS (-87%) |
| Pixel Rate | 225.6 GPixel/s | 483.8 GPixel/s | 258.2 GPixel/s (114%) |
| Texture Rate | 609.1 GTexel/s | 1290 GTexel/s | 680.9 GTexel/s (112%) |
Price
| NVIDIA A100 40 GB (PCIe) | NVIDIA RTX 4090 | Difference | |
|---|---|---|---|
| Release Date | Jun 22nd, 2020 | Oct 12th, 2022 | |
| MSRP | - | $1,599.00 |
Test bench configuration
| NVIDIA A100 40 GB (PCIe) | NVIDIA RTX 4090 | Difference | |
|---|---|---|---|
| Hardware | BIZON X5000 More details | BIZON X5500 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: Nvidia Driver: VRay Benchmark: Octane Benchmark: Redshift Benchmark: Blender: Luxmark: |
Recommended hardware
| NVIDIA A100 40 GB (PCIe) | NVIDIA RTX 4090 | Difference | |
|---|---|---|---|
| Best GPU workstations | BIZON X5500 | BIZON X5500 | - |
| Best NVIDIA GPU servers | BIZON X7000 | - |
