GPU Benchmarks NVIDIA Tesla V100 vs. NVIDIA A100 40 GB (PCIe) vs. NVIDIA H100 NVL (PCIe) vs. NVIDIA RTX 4090
NVIDIA Tesla V100
NVIDIA A100 40 GB (PCIe)
NVIDIA H100 NVL (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 Tesla V100 vs NVIDIA A100 40 GB (PCIe) vs NVIDIA H100 NVL (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 April 2025
Resnet50 (FP16)
1 GPU
NVIDIA Tesla V100
706.07 points
NVIDIA A100 40 GB (PCIe)
2179 points
NVIDIA H100 NVL (PCIe)
3042 points
NVIDIA RTX 4090
1720 points
4 GPU
NVIDIA Tesla V100
2309.02 points
NVIDIA A100 40 GB (PCIe)
8561 points
NVIDIA H100 NVL (PCIe)
11989 points
NVIDIA RTX 4090
5934 points
8 GPU
NVIDIA Tesla V100
n/a
NVIDIA A100 40 GB (PCIe)
16797 points
NVIDIA H100 NVL (PCIe)
30070 points
NVIDIA RTX 4090
n/a
higher is better
Resnet50 (FP16 XLA)
1 GPU
NVIDIA Tesla V100
1071.32 points
NVIDIA A100 40 GB (PCIe)
n/a
NVIDIA H100 NVL (PCIe)
n/a
NVIDIA RTX 4090
n/a
4 GPU
NVIDIA Tesla V100
3562.89 points
NVIDIA A100 40 GB (PCIe)
n/a
NVIDIA H100 NVL (PCIe)
n/a
NVIDIA RTX 4090
n/a
higher is better
Resnet50 (FP32)
1 GPU
NVIDIA Tesla V100
321.57 points
NVIDIA A100 40 GB (PCIe)
1001 points
NVIDIA H100 NVL (PCIe)
1350 points
NVIDIA RTX 4090
927 points
4 GPU
NVIDIA Tesla V100
1141.56 points
NVIDIA A100 40 GB (PCIe)
3849 points
NVIDIA H100 NVL (PCIe)
5513 points
NVIDIA RTX 4090
1715 points
8 GPU
NVIDIA Tesla V100
n/a
NVIDIA A100 40 GB (PCIe)
7557 points
NVIDIA H100 NVL (PCIe)
14725 points
NVIDIA RTX 4090
n/a
higher is better
Resnet50 (FP32 XLA)
1 GPU
NVIDIA Tesla V100
351.14 points
NVIDIA A100 40 GB (PCIe)
n/a
NVIDIA H100 NVL (PCIe)
n/a
NVIDIA RTX 4090
n/a
4 GPU
NVIDIA Tesla V100
1202.04 points
NVIDIA A100 40 GB (PCIe)
n/a
NVIDIA H100 NVL (PCIe)
n/a
NVIDIA RTX 4090
n/a
higher is better
Resnet152 (FP16)
1 GPU
NVIDIA Tesla V100
308.35 points
NVIDIA A100 40 GB (PCIe)
930 points
NVIDIA H100 NVL (PCIe)
1232 points
NVIDIA RTX 4090
n/a
4 GPU
NVIDIA Tesla V100
984.5 points
NVIDIA A100 40 GB (PCIe)
3557 points
NVIDIA H100 NVL (PCIe)
4858 points
NVIDIA RTX 4090
n/a
8 GPU
NVIDIA Tesla V100
n/a
NVIDIA A100 40 GB (PCIe)
6809 points
NVIDIA H100 NVL (PCIe)
12411 points
NVIDIA RTX 4090
n/a
higher is better
Resnet152 (FP16 XLA)
1 GPU
NVIDIA Tesla V100
419.46 points
NVIDIA A100 40 GB (PCIe)
n/a
NVIDIA H100 NVL (PCIe)
n/a
NVIDIA RTX 4090
n/a
4 GPU
NVIDIA Tesla V100
1412.1 points
NVIDIA A100 40 GB (PCIe)
n/a
NVIDIA H100 NVL (PCIe)
n/a
NVIDIA RTX 4090
n/a
higher is better
Resnet152 (FP32)
1 GPU
NVIDIA Tesla V100
134.94 points
NVIDIA A100 40 GB (PCIe)
409 points
NVIDIA H100 NVL (PCIe)
520 points
NVIDIA RTX 4090
n/a
4 GPU
NVIDIA Tesla V100
452.05 points
NVIDIA A100 40 GB (PCIe)
1498 points
NVIDIA H100 NVL (PCIe)
2124 points
NVIDIA RTX 4090
n/a
8 GPU
NVIDIA Tesla V100
n/a
NVIDIA A100 40 GB (PCIe)
2851 points
NVIDIA H100 NVL (PCIe)
5950 points
NVIDIA RTX 4090
n/a
higher is better
Resnet152 (FP32 XLA)
1 GPU
NVIDIA Tesla V100
143.28 points
NVIDIA A100 40 GB (PCIe)
n/a
NVIDIA H100 NVL (PCIe)
n/a
NVIDIA RTX 4090
n/a
4 GPU
NVIDIA Tesla V100
519.5 points
NVIDIA A100 40 GB (PCIe)
n/a
NVIDIA H100 NVL (PCIe)
n/a
NVIDIA RTX 4090
n/a
higher is better
Inception V3 (FP16)
1 GPU
NVIDIA Tesla V100
424.55 points
NVIDIA A100 40 GB (PCIe)
1283 points
NVIDIA H100 NVL (PCIe)
1835 points
NVIDIA RTX 4090
n/a
4 GPU
NVIDIA Tesla V100
1490.05 points
NVIDIA A100 40 GB (PCIe)
5218 points
NVIDIA H100 NVL (PCIe)
7232 points
NVIDIA RTX 4090
n/a
8 GPU
NVIDIA Tesla V100
n/a
NVIDIA A100 40 GB (PCIe)
10122 points
NVIDIA H100 NVL (PCIe)
15827 points
NVIDIA RTX 4090
n/a
higher is better
Inception V3 (FP16 XLA)
1 GPU
NVIDIA Tesla V100
585.31 points
NVIDIA A100 40 GB (PCIe)
n/a
NVIDIA H100 NVL (PCIe)
n/a
NVIDIA RTX 4090
n/a
4 GPU
NVIDIA Tesla V100
2121.59 points
NVIDIA A100 40 GB (PCIe)
n/a
NVIDIA H100 NVL (PCIe)
n/a
NVIDIA RTX 4090
n/a
higher is better
Inception V3 (FP32)
1 GPU
NVIDIA Tesla V100
211.08 points
NVIDIA A100 40 GB (PCIe)
658 points
NVIDIA H100 NVL (PCIe)
854 points
NVIDIA RTX 4090
n/a
4 GPU
NVIDIA Tesla V100
785.13 points
NVIDIA A100 40 GB (PCIe)
2568 points
NVIDIA H100 NVL (PCIe)
3487 points
NVIDIA RTX 4090
n/a
8 GPU
NVIDIA Tesla V100
n/a
NVIDIA A100 40 GB (PCIe)
5058 points
NVIDIA H100 NVL (PCIe)
9837 points
NVIDIA RTX 4090
n/a
higher is better
Inception V3 (FP32 XLA)
1 GPU
NVIDIA Tesla V100
228.85 points
NVIDIA A100 40 GB (PCIe)
n/a
NVIDIA H100 NVL (PCIe)
n/a
NVIDIA RTX 4090
n/a
4 GPU
NVIDIA Tesla V100
820.35 points
NVIDIA A100 40 GB (PCIe)
n/a
NVIDIA H100 NVL (PCIe)
n/a
NVIDIA RTX 4090
n/a
higher is better
Inception V4 (FP16)
1 GPU
NVIDIA Tesla V100
225.21 points
NVIDIA A100 40 GB (PCIe)
616 points
NVIDIA H100 NVL (PCIe)
848 points
NVIDIA RTX 4090
n/a
4 GPU
NVIDIA Tesla V100
816.02 points
NVIDIA A100 40 GB (PCIe)
2377 points
NVIDIA H100 NVL (PCIe)
3343 points
NVIDIA RTX 4090
n/a
8 GPU
NVIDIA Tesla V100
n/a
NVIDIA A100 40 GB (PCIe)
4532 points
NVIDIA H100 NVL (PCIe)
6819 points
NVIDIA RTX 4090
n/a
higher is better
Inception V4 (FP16 XLA)
1 GPU
NVIDIA Tesla V100
318.96 points
NVIDIA A100 40 GB (PCIe)
n/a
NVIDIA H100 NVL (PCIe)
n/a
NVIDIA RTX 4090
n/a
4 GPU
NVIDIA Tesla V100
1142.94 points
NVIDIA A100 40 GB (PCIe)
n/a
NVIDIA H100 NVL (PCIe)
n/a
NVIDIA RTX 4090
n/a
higher is better
Inception V4 (FP32)
1 GPU
NVIDIA Tesla V100
102.53 points
NVIDIA A100 40 GB (PCIe)
290 points
NVIDIA H100 NVL (PCIe)
380 points
NVIDIA RTX 4090
n/a
4 GPU
NVIDIA Tesla V100
383.89 points
NVIDIA A100 40 GB (PCIe)
1031 points
NVIDIA H100 NVL (PCIe)
1551 points
NVIDIA RTX 4090
n/a
8 GPU
NVIDIA Tesla V100
n/a
NVIDIA A100 40 GB (PCIe)
1950 points
NVIDIA H100 NVL (PCIe)
4496 points
NVIDIA RTX 4090
n/a
higher is better
Inception V4 (FP32 XLA)
1 GPU
NVIDIA Tesla V100
108.26 points
NVIDIA A100 40 GB (PCIe)
n/a
NVIDIA H100 NVL (PCIe)
n/a
NVIDIA RTX 4090
n/a
4 GPU
NVIDIA Tesla V100
415.26 points
NVIDIA A100 40 GB (PCIe)
n/a
NVIDIA H100 NVL (PCIe)
n/a
NVIDIA RTX 4090
n/a
higher is better
VGG16 (FP16)
1 GPU
NVIDIA Tesla V100
381.74 points
NVIDIA A100 40 GB (PCIe)
1249 points
NVIDIA H100 NVL (PCIe)
1085 points
NVIDIA RTX 4090
n/a
4 GPU
NVIDIA Tesla V100
835.56 points
NVIDIA A100 40 GB (PCIe)
4989 points
NVIDIA H100 NVL (PCIe)
4276 points
NVIDIA RTX 4090
n/a
8 GPU
NVIDIA Tesla V100
n/a
NVIDIA A100 40 GB (PCIe)
10733 points
NVIDIA H100 NVL (PCIe)
15908 points
NVIDIA RTX 4090
n/a
higher is better
VGG16 (FP16 XLA)
1 GPU
NVIDIA Tesla V100
469.5 points
NVIDIA A100 40 GB (PCIe)
n/a
NVIDIA H100 NVL (PCIe)
n/a
NVIDIA RTX 4090
n/a
4 GPU
NVIDIA Tesla V100
1338.8 points
NVIDIA A100 40 GB (PCIe)
n/a
NVIDIA H100 NVL (PCIe)
n/a
NVIDIA RTX 4090
n/a
higher is better
VGG16 (FP32)
1 GPU
NVIDIA Tesla V100
189.68 points
NVIDIA A100 40 GB (PCIe)
529 points
NVIDIA H100 NVL (PCIe)
776 points
NVIDIA RTX 4090
n/a
4 GPU
NVIDIA Tesla V100
480.12 points
NVIDIA A100 40 GB (PCIe)
2215 points
NVIDIA H100 NVL (PCIe)
3171 points
NVIDIA RTX 4090
n/a
8 GPU
NVIDIA Tesla V100
n/a
NVIDIA A100 40 GB (PCIe)
4278 points
NVIDIA H100 NVL (PCIe)
8762 points
NVIDIA RTX 4090
n/a
higher is better
VGG16 (FP32 XLA)
1 GPU
NVIDIA Tesla V100
206.92 points
NVIDIA A100 40 GB (PCIe)
n/a
NVIDIA H100 NVL (PCIe)
n/a
NVIDIA RTX 4090
n/a
4 GPU
NVIDIA Tesla V100
560.94 points
NVIDIA A100 40 GB (PCIe)
n/a
NVIDIA H100 NVL (PCIe)
n/a
NVIDIA RTX 4090
n/a
higher is better
3D, GPU Rendering Benchmarks 2024–2025 [Updated]↑
As of April 2025
V-Ray
1 GPU
NVIDIA Tesla V100
505 points
NVIDIA A100 40 GB (PCIe)
1555 points
NVIDIA H100 NVL (PCIe)
n/a
NVIDIA RTX 4090
5556 points
higher is better
Octane
1 GPU
NVIDIA Tesla V100
356 points
NVIDIA A100 40 GB (PCIe)
498 points
NVIDIA H100 NVL (PCIe)
n/a
NVIDIA RTX 4090
1445 points
higher is better
Redshift
1 GPU
NVIDIA Tesla V100
8.24 minutes
NVIDIA A100 40 GB (PCIe)
n/a
NVIDIA H100 NVL (PCIe)
n/a
NVIDIA RTX 4090
1.16 minutes
lower is better
Blender
1 GPU
NVIDIA Tesla V100
2075.51 score
NVIDIA A100 40 GB (PCIe)
3788 score
NVIDIA H100 NVL (PCIe)
5069.86 score
NVIDIA RTX 4090
12123.96 score
higher is better
Luxmark
1 GPU
NVIDIA Tesla V100
35698 points
NVIDIA A100 40 GB (PCIe)
n/a
NVIDIA H100 NVL (PCIe)
n/a
NVIDIA RTX 4090
158815 points
higher is better
Unreal Engine
1 GPU
NVIDIA Tesla V100
n/a
NVIDIA A100 40 GB (PCIe)
n/a
NVIDIA H100 NVL (PCIe)
n/a
NVIDIA RTX 4090
92.1 FPS
higher is better
RELION Cryo-EM Benchmarks 2024-2025 [Updated]↑
As of April 2025
Total run time
1 GPU
NVIDIA Tesla V100
n/a
NVIDIA A100 40 GB (PCIe)
178.9 Min
NVIDIA H100 NVL (PCIe)
n/a
NVIDIA RTX 4090
105.2 Min
4 GPU
NVIDIA Tesla V100
n/a
NVIDIA A100 40 GB (PCIe)
50.3 Min
NVIDIA H100 NVL (PCIe)
n/a
NVIDIA RTX 4090
53.2 Min
lower is better
Llama3 70B Inference Benchmark 2024–2025 [Updated]↑
As of April 2025
Eval rate
1 GPU
NVIDIA Tesla V100
n/a
NVIDIA A100 40 GB (PCIe)
n/a
NVIDIA H100 NVL (PCIe)
n/a
NVIDIA RTX 4090
9.95 tokens/s
2 GPU
NVIDIA Tesla V100
n/a
NVIDIA A100 40 GB (PCIe)
n/a
NVIDIA H100 NVL (PCIe)
n/a
NVIDIA RTX 4090
19.99 tokens/s
higher is better
Technical Specifications
Board Design
| NVIDIA Tesla V100 | NVIDIA A100 40 GB (PCIe) | NVIDIA H100 NVL (PCIe) | NVIDIA RTX 4090 | |
|---|---|---|---|---|
| Length | 11 in / 267 mm | 11 in / 267 mm | 11 in / 268 mm | 13 in / 336 mm |
| Outputs | No outputs | No outputs | No outputs | 1x HDMI, 3x DisplayPort |
| Power Connectors | 2x 8-pin | 8-pin EPS | 8-pin EPS | 1x 16-pin |
| Slot width | Dual-slot | Dual-slot | Dual-slot | Triple-slot |
| TDP | 250 W | 250 W | 700 W | 450 W |
Clock Speeds
| NVIDIA Tesla V100 | NVIDIA A100 40 GB (PCIe) | NVIDIA H100 NVL (PCIe) | NVIDIA RTX 4090 | |
|---|---|---|---|---|
| Boost Clock | 1380 MHz | 1410 MHz | 1837 MHz | 2520 MHz |
| GPU Clock | 1230 MHz | 765 MHz | 1665 MHz | 2235 MHz |
| Memory Clock | 1752 MHz | 2400 MHz | 5300 MHz | 21200 MHz |
Graphics Card
| NVIDIA Tesla V100 | NVIDIA A100 40 GB (PCIe) | NVIDIA H100 NVL (PCIe) | NVIDIA RTX 4090 | |
|---|---|---|---|---|
| Bus Interface | PCIe 3.0 x16 | PCIe 4.0 x16 | PCIe 5.0 x16 | PCIe 4.0 x16 |
| Generation | Tesla (Vxx) | Tesla (Axx) | Server Hopper (Hxx) | GeForce 40 |
Graphics Features
| NVIDIA Tesla V100 | NVIDIA A100 40 GB (PCIe) | NVIDIA H100 NVL (PCIe) | NVIDIA RTX 4090 | |
|---|---|---|---|---|
| DirectX | 12 Ultimate (12_1) | - | - | 12 Ultimate (12_2) |
| OpenCL | 1.2 | 2 | 3 | 3 |
| OpenGL | 4.6 | - | - | 4.6 |
| Shader Model | 6.4 | - | - | 6.7 |
| CUDA | 7 | 8 | 9 | 8.9 |
Graphics Processor
| NVIDIA Tesla V100 | NVIDIA A100 40 GB (PCIe) | NVIDIA H100 NVL (PCIe) | NVIDIA RTX 4090 | |
|---|---|---|---|---|
| Architecture | Volta | Ampere | Hopper | Ada Lovelace |
| Die Size | 815 mm2 | 826 mm2 | 814 mm2 | 608 mm2 |
| GPU Name | GV100 | GA100 | GH100 | AD102-300-A1 |
| Process Size | 12 nm | 7 nm | 5 nm | 5 nm |
| Transistors | 21100 million | 54200 million | 80000 million | 76300 million |
Memory
| NVIDIA Tesla V100 | NVIDIA A100 40 GB (PCIe) | NVIDIA H100 NVL (PCIe) | NVIDIA RTX 4090 | |
|---|---|---|---|---|
| Bandwidth | 897 GB/s | 1555 GB/s | 3360 GB/s | 1018 GB/s |
| Memory Bus | 4096 bit | 5120 bit | 5120 bit | 384 bit |
| Memory Size | 32 GB | 40 GB | 96 GB | 24 GB |
| Memory Type | HBM2 | HBM2e | HBM3 | GDDR6X |
Render Config
| NVIDIA Tesla V100 | NVIDIA A100 40 GB (PCIe) | NVIDIA H100 NVL (PCIe) | NVIDIA RTX 4090 | |
|---|---|---|---|---|
| ROPs | 128 | 160 | 24 | 192 |
| Shading Units/ CUDA Cores | 5120 | 6912 | 16896 | 16384 |
| TMUs | 320 | 432 | 528 | 512 |
| Tensor Cores | 640 | 432 | 528 | 512 |
| RT Cores | - | - | - | 128 |
Theoretical Performance
| NVIDIA Tesla V100 | NVIDIA A100 40 GB (PCIe) | NVIDIA H100 NVL (PCIe) | NVIDIA RTX 4090 | |
|---|---|---|---|---|
| FP16 (half) performance | 28.26 TFLOPS | 77.97 TFLOPS | 248.3 TFLOPS | 82.58 TFLOPS |
| FP32 (float) performance | 14.13 TFLOPS | 19.49 TFLOPS | 62.08 TFLOPS | 82.58 TFLOPS |
| FP64 (double) performance | 7066 GFLOPS | 9746 GFLOPS | 31040 GFLOPS | 1290 GFLOPS |
| Pixel Rate | 176.6 GPixel/s | 225.6 GPixel/s | 44.09 GPixel/s | 483.8 GPixel/s |
| Texture Rate | 441.6 GTexel/s | 609.1 GTexel/s | 969.9 GTexel/s | 1290 GTexel/s |
Price
| NVIDIA Tesla V100 | NVIDIA A100 40 GB (PCIe) | NVIDIA H100 NVL (PCIe) | NVIDIA RTX 4090 | |
|---|---|---|---|---|
| MSRP | $11,459.00 | - | - | $1,599.00 |
| Release Date | Mar 27th, 2018 | Jun 22nd, 2020 | Mar 21st, 2023 | Oct 12th, 2022 |
Test bench configuration
| NVIDIA Tesla V100 | NVIDIA A100 40 GB (PCIe) | NVIDIA H100 NVL (PCIe) | NVIDIA RTX 4090 | |
|---|---|---|---|---|
| Hardware | BIZON X5000 More details | BIZON X5000 More details | BIZON X5000 More details | BIZON X5500 More details |
| Software | Deep learning: Nvidia Driver: 440 CUDA: 10.1 TensorFlow: 1.14 Batch size: 64 3D Rendering: Nvidia Driver: 442.19 VRay Benchmark: 4.10.3 Octane Benchmark: 4.00 Redshift Benchmark: 3.0.x Blender: 2.81 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 | 3D Rendering: 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 Tesla V100 | NVIDIA A100 40 GB (PCIe) | NVIDIA H100 NVL (PCIe) | NVIDIA RTX 4090 | |
|---|---|---|---|---|
| Best GPU workstations | - | BIZON X5500 | BIZON X5500 | BIZON X5500 |
| Best NVIDIA GPU servers | - | BIZON X7000 | BIZON X7000 | - |
