GPU Benchmarks NVIDIA RTX 3090 vs. NVIDIA A100 40 GB (PCIe)
NVIDIA RTX 3090
NVIDIA A100 40 GB (PCIe)
Quick links:
- Benchmarks
- Specifications
- Deep Learning GPU Benchmarks 2024–2025
- 3D, GPU Rendering Benchmarks 2024–2025
- RELION Cryo-EM 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 RTX 3090 vs NVIDIA A100 40 GB (PCIe) 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 April 2025
Resnet50 (FP16)
1 GPU
NVIDIA RTX 3090
1071 points
NVIDIA A100 40 GB (PCIe)
2179 points
4 GPU
NVIDIA RTX 3090
2922 points
NVIDIA A100 40 GB (PCIe)
8561 points
8 GPU
NVIDIA RTX 3090
n/a
NVIDIA A100 40 GB (PCIe)
16797 points
higher is better
Resnet50 (FP32)
1 GPU
NVIDIA RTX 3090
596 points
NVIDIA A100 40 GB (PCIe)
1001 points
4 GPU
NVIDIA RTX 3090
1625 points
NVIDIA A100 40 GB (PCIe)
3849 points
8 GPU
NVIDIA RTX 3090
n/a
NVIDIA A100 40 GB (PCIe)
7557 points
higher is better
Resnet152 (FP16)
1 GPU
NVIDIA RTX 3090
491 points
NVIDIA A100 40 GB (PCIe)
930 points
4 GPU
NVIDIA RTX 3090
1029 points
NVIDIA A100 40 GB (PCIe)
3557 points
8 GPU
NVIDIA RTX 3090
n/a
NVIDIA A100 40 GB (PCIe)
6809 points
higher is better
Resnet152 (FP32)
1 GPU
NVIDIA RTX 3090
223 points
NVIDIA A100 40 GB (PCIe)
409 points
4 GPU
NVIDIA RTX 3090
434 points
NVIDIA A100 40 GB (PCIe)
1498 points
8 GPU
NVIDIA RTX 3090
n/a
NVIDIA A100 40 GB (PCIe)
2851 points
higher is better
Inception V3 (FP16)
1 GPU
NVIDIA RTX 3090
715 points
NVIDIA A100 40 GB (PCIe)
1283 points
4 GPU
NVIDIA RTX 3090
1720 points
NVIDIA A100 40 GB (PCIe)
5218 points
8 GPU
NVIDIA RTX 3090
n/a
NVIDIA A100 40 GB (PCIe)
10122 points
higher is better
Inception V3 (FP32)
1 GPU
NVIDIA RTX 3090
399 points
NVIDIA A100 40 GB (PCIe)
658 points
4 GPU
NVIDIA RTX 3090
1267 points
NVIDIA A100 40 GB (PCIe)
2568 points
8 GPU
NVIDIA RTX 3090
n/a
NVIDIA A100 40 GB (PCIe)
5058 points
higher is better
Inception V4 (FP16)
1 GPU
NVIDIA RTX 3090
383 points
NVIDIA A100 40 GB (PCIe)
616 points
4 GPU
NVIDIA RTX 3090
1022 points
NVIDIA A100 40 GB (PCIe)
2377 points
8 GPU
NVIDIA RTX 3090
n/a
NVIDIA A100 40 GB (PCIe)
4532 points
higher is better
Inception V4 (FP32)
1 GPU
NVIDIA RTX 3090
156 points
NVIDIA A100 40 GB (PCIe)
290 points
4 GPU
NVIDIA RTX 3090
319 points
NVIDIA A100 40 GB (PCIe)
1031 points
8 GPU
NVIDIA RTX 3090
n/a
NVIDIA A100 40 GB (PCIe)
1950 points
higher is better
VGG16 (FP16)
1 GPU
NVIDIA RTX 3090
577 points
NVIDIA A100 40 GB (PCIe)
1249 points
4 GPU
NVIDIA RTX 3090
1004 points
NVIDIA A100 40 GB (PCIe)
4989 points
8 GPU
NVIDIA RTX 3090
n/a
NVIDIA A100 40 GB (PCIe)
10733 points
higher is better
VGG16 (FP32)
1 GPU
NVIDIA RTX 3090
341 points
NVIDIA A100 40 GB (PCIe)
529 points
4 GPU
NVIDIA RTX 3090
1077 points
NVIDIA A100 40 GB (PCIe)
2215 points
8 GPU
NVIDIA RTX 3090
n/a
NVIDIA A100 40 GB (PCIe)
4278 points
higher is better
3D, GPU Rendering Benchmarks 2024–2025 [Updated]↑
As of April 2025
V-Ray
1 GPU
NVIDIA RTX 3090
2185 points
NVIDIA A100 40 GB (PCIe)
1555 points
higher is better
Octane
1 GPU
NVIDIA RTX 3090
769 points
NVIDIA A100 40 GB (PCIe)
498 points
higher is better
Redshift
1 GPU
NVIDIA RTX 3090
2.3 minutes
NVIDIA A100 40 GB (PCIe)
n/a
lower is better
Blender
1 GPU
NVIDIA RTX 3090
6010.36 score
NVIDIA A100 40 GB (PCIe)
3788 score
higher is better
Luxmark
1 GPU
NVIDIA RTX 3090
95123 points
NVIDIA A100 40 GB (PCIe)
n/a
higher is better
Unreal Engine
1 GPU
NVIDIA RTX 3090
47.6 FPS
NVIDIA A100 40 GB (PCIe)
n/a
higher is better
RELION Cryo-EM Benchmarks 2024-2025 [Updated]↑
As of April 2025
Total run time
1 GPU
NVIDIA RTX 3090
165.9 Min
NVIDIA A100 40 GB (PCIe)
178.9 Min
4 GPU
NVIDIA RTX 3090
57.1 Min
NVIDIA A100 40 GB (PCIe)
50.3 Min
lower is better
Technical Specifications
Board Design
| NVIDIA RTX 3090 | NVIDIA A100 40 GB (PCIe) | Difference | |
|---|---|---|---|
| Length | 13 in / 336 mm | 11 in / 267 mm | -3 in / -69 mm (-21%) |
| Outputs | 1x HDMI, 3x DisplayPort | No outputs | |
| Power Connectors | 1x 12-pin | 8-pin EPS | |
| Slot width | Triple-slot | Dual-slot | |
| TDP | 350 W | 250 W | -100 W (-29%) |
Clock Speeds
| NVIDIA RTX 3090 | NVIDIA A100 40 GB (PCIe) | Difference | |
|---|---|---|---|
| Boost Clock | 1695 MHz | 1410 MHz | -285 MHz (-17%) |
| GPU Clock | 1395 MHz | 765 MHz | -630 MHz (-45%) |
| Memory Clock | 19500 MHz | 2400 MHz | -17100 MHz (-88%) |
Graphics Card
| NVIDIA RTX 3090 | NVIDIA A100 40 GB (PCIe) | Difference | |
|---|---|---|---|
| Bus Interface | PCIe 4.0 x16 | PCIe 4.0 x16 | - |
| Generation | GeForce 30 | Tesla (Axx) |
Graphics Features
| NVIDIA RTX 3090 | NVIDIA A100 40 GB (PCIe) | Difference | |
|---|---|---|---|
| DirectX | 12 Ultimate (12_2) | - | |
| OpenCL | 3 | 2 | |
| OpenGL | 4.6 | - | |
| Shader Model | 6.6 | - | |
| CUDA | 8.6 | 8 |
Graphics Processor
| NVIDIA RTX 3090 | NVIDIA A100 40 GB (PCIe) | Difference | |
|---|---|---|---|
| Architecture | Ampere | Ampere | - |
| Die Size | 628 mm2 | 826 mm2 | 198 mm2 (32%) |
| GPU Name | GA102-300-A1 | GA100 | |
| Process Size | 8 nm | 7 nm | -1 nm (-12%) |
| Transistors | 28300 million | 54200 million | 25900 million (92%) |
Memory
| NVIDIA RTX 3090 | NVIDIA A100 40 GB (PCIe) | Difference | |
|---|---|---|---|
| Bandwidth | 936.2 GB/s | 1555 GB/s | 618.8 GB/s (66%) |
| Memory Bus | 384 bit | 5120 bit | 4736 bit (1233%) |
| Memory Size | 24 GB | 40 GB | 16 GB (67%) |
| Memory Type | GDDR6X | HBM2e |
Render Config
| NVIDIA RTX 3090 | NVIDIA A100 40 GB (PCIe) | Difference | |
|---|---|---|---|
| ROPs | 112 | 160 | 48 (43%) |
| RT Cores | 82 | - | |
| Shading Units/ CUDA Cores | 10496 | 6912 | -3584 (-34%) |
| TMUs | 328 | 432 | 104 (32%) |
| Tensor Cores | 328 | 432 | 104 (32%) |
Theoretical Performance
| NVIDIA RTX 3090 | NVIDIA A100 40 GB (PCIe) | Difference | |
|---|---|---|---|
| FP16 (half) performance | 35.58 TFLOPS | 77.97 TFLOPS | 42.39 TFLOPS (119%) |
| FP32 (float) performance | 35.58 TFLOPS | 19.49 TFLOPS | -16.09 TFLOPS (-45%) |
| FP64 (double) performance | 556 GFLOPS | 9746 GFLOPS | 9190 GFLOPS (1653%) |
| Pixel Rate | 189.8 GPixel/s | 225.6 GPixel/s | 35.8 GPixel/s (19%) |
| Texture Rate | 556 GTexel/s | 609.1 GTexel/s | 53.1 GTexel/s (10%) |
Price
| NVIDIA RTX 3090 | NVIDIA A100 40 GB (PCIe) | Difference | |
|---|---|---|---|
| MSRP | $1,499.00 | - | |
| Release Date | Sep 1st, 2020 | Jun 22nd, 2020 |
Test bench configuration
| NVIDIA RTX 3090 | NVIDIA A100 40 GB (PCIe) | Difference | |
|---|---|---|---|
| Hardware | BIZON X5000 More details | BIZON X5000 More details | - |
| Software | 3D Rendering: Nvidia Driver: 456.38 VRay Benchmark: 5 Octane Benchmark: 2020.1.5 Redshift Benchmark: 3.0.28 Demo Blender: 2.90 Luxmark: 3.1 | 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 |
Recommended hardware
| NVIDIA RTX 3090 | NVIDIA A100 40 GB (PCIe) | Difference | |
|---|---|---|---|
| Best GPU workstations | BIZON X5500 | BIZON X5500 | |
| Best NVIDIA GPU servers | - | BIZON X7000 |
