Compare
NVIDIA RTX 4090 vs NVIDIA L40
Quick links:
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 these 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]
Resnet50 (FP16)
1 GPU
NVIDIA RTX 4090
1720
NVIDIA L40
n/a
4 GPU
NVIDIA RTX 4090
5934
NVIDIA L40
n/a
8 GPU
NVIDIA RTX 4090
n/a
NVIDIA L40
n/a
Resnet50 (FP16 XLA)
1 GPU
NVIDIA RTX 4090
n/a
NVIDIA L40
n/a
4 GPU
NVIDIA RTX 4090
n/a
NVIDIA L40
n/a
8 GPU
NVIDIA RTX 4090
n/a
NVIDIA L40
n/a
Resnet50 (FP32)
1 GPU
NVIDIA RTX 4090
927
NVIDIA L40
n/a
4 GPU
NVIDIA RTX 4090
1715
NVIDIA L40
n/a
8 GPU
NVIDIA RTX 4090
n/a
NVIDIA L40
n/a
Resnet50 (FP32 XLA)
1 GPU
NVIDIA RTX 4090
n/a
NVIDIA L40
n/a
4 GPU
NVIDIA RTX 4090
n/a
NVIDIA L40
n/a
8 GPU
NVIDIA RTX 4090
n/a
NVIDIA L40
n/a
Resnet152 (FP16)
1 GPU
NVIDIA RTX 4090
n/a
NVIDIA L40
n/a
4 GPU
NVIDIA RTX 4090
n/a
NVIDIA L40
n/a
8 GPU
NVIDIA RTX 4090
n/a
NVIDIA L40
n/a
Resnet152 (FP16 XLA)
1 GPU
NVIDIA RTX 4090
n/a
NVIDIA L40
n/a
4 GPU
NVIDIA RTX 4090
n/a
NVIDIA L40
n/a
8 GPU
NVIDIA RTX 4090
n/a
NVIDIA L40
n/a
Resnet152 (FP32)
1 GPU
NVIDIA RTX 4090
n/a
NVIDIA L40
n/a
4 GPU
NVIDIA RTX 4090
n/a
NVIDIA L40
n/a
8 GPU
NVIDIA RTX 4090
n/a
NVIDIA L40
n/a
Resnet152 (FP32 XLA)
1 GPU
NVIDIA RTX 4090
n/a
NVIDIA L40
n/a
4 GPU
NVIDIA RTX 4090
n/a
NVIDIA L40
n/a
8 GPU
NVIDIA RTX 4090
n/a
NVIDIA L40
n/a
Inception V3 (FP16)
1 GPU
NVIDIA RTX 4090
n/a
NVIDIA L40
n/a
4 GPU
NVIDIA RTX 4090
n/a
NVIDIA L40
n/a
8 GPU
NVIDIA RTX 4090
n/a
NVIDIA L40
n/a
Inception V3 (FP16 XLA)
1 GPU
NVIDIA RTX 4090
n/a
NVIDIA L40
n/a
4 GPU
NVIDIA RTX 4090
n/a
NVIDIA L40
n/a
8 GPU
NVIDIA RTX 4090
n/a
NVIDIA L40
n/a
Inception V3 (FP32)
1 GPU
NVIDIA RTX 4090
n/a
NVIDIA L40
n/a
4 GPU
NVIDIA RTX 4090
n/a
NVIDIA L40
n/a
8 GPU
NVIDIA RTX 4090
n/a
NVIDIA L40
n/a
Inception V3 (FP32 XLA)
1 GPU
NVIDIA RTX 4090
n/a
NVIDIA L40
n/a
4 GPU
NVIDIA RTX 4090
n/a
NVIDIA L40
n/a
8 GPU
NVIDIA RTX 4090
n/a
NVIDIA L40
n/a
Inception V4 (FP16)
1 GPU
NVIDIA RTX 4090
n/a
NVIDIA L40
n/a
4 GPU
NVIDIA RTX 4090
n/a
NVIDIA L40
n/a
8 GPU
NVIDIA RTX 4090
n/a
NVIDIA L40
n/a
Inception V4 (FP16 XLA)
1 GPU
NVIDIA RTX 4090
n/a
NVIDIA L40
n/a
4 GPU
NVIDIA RTX 4090
n/a
NVIDIA L40
n/a
8 GPU
NVIDIA RTX 4090
n/a
NVIDIA L40
n/a
Inception V4 (FP32)
1 GPU
NVIDIA RTX 4090
n/a
NVIDIA L40
n/a
4 GPU
NVIDIA RTX 4090
n/a
NVIDIA L40
n/a
8 GPU
NVIDIA RTX 4090
n/a
NVIDIA L40
n/a
Inception V4 (FP32 XLA)
1 GPU
NVIDIA RTX 4090
n/a
NVIDIA L40
n/a
4 GPU
NVIDIA RTX 4090
n/a
NVIDIA L40
n/a
8 GPU
NVIDIA RTX 4090
n/a
NVIDIA L40
n/a
VGG16 (FP16)
1 GPU
NVIDIA RTX 4090
n/a
NVIDIA L40
n/a
4 GPU
NVIDIA RTX 4090
n/a
NVIDIA L40
n/a
8 GPU
NVIDIA RTX 4090
n/a
NVIDIA L40
n/a
VGG16 (FP16 XLA)
1 GPU
NVIDIA RTX 4090
n/a
NVIDIA L40
n/a
4 GPU
NVIDIA RTX 4090
n/a
NVIDIA L40
n/a
8 GPU
NVIDIA RTX 4090
n/a
NVIDIA L40
n/a
VGG16 (FP32)
1 GPU
NVIDIA RTX 4090
n/a
NVIDIA L40
n/a
4 GPU
NVIDIA RTX 4090
n/a
NVIDIA L40
n/a
8 GPU
NVIDIA RTX 4090
n/a
NVIDIA L40
n/a
VGG16 (FP32 XLA)
1 GPU
NVIDIA RTX 4090
n/a
NVIDIA L40
n/a
4 GPU
NVIDIA RTX 4090
n/a
NVIDIA L40
n/a
8 GPU
NVIDIA RTX 4090
n/a
NVIDIA L40
n/a
Googlenet (FP16)
1 GPU
NVIDIA RTX 4090
n/a
NVIDIA L40
n/a
4 GPU
NVIDIA RTX 4090
n/a
NVIDIA L40
n/a
8 GPU
NVIDIA RTX 4090
n/a
NVIDIA L40
n/a
Googlenet (FP32)
1 GPU
NVIDIA RTX 4090
n/a
NVIDIA L40
n/a
4 GPU
NVIDIA RTX 4090
n/a
NVIDIA L40
n/a
8 GPU
NVIDIA RTX 4090
n/a
NVIDIA L40
n/a
3D, GPU Rendering Benchmarks 2024–2025 [Updated]
V-Ray
1 GPU
NVIDIA RTX 4090
5556
NVIDIA L40
n/a
4 GPU
NVIDIA RTX 4090
n/a
NVIDIA L40
n/a
8 GPU
NVIDIA RTX 4090
n/a
NVIDIA L40
n/a
Octane
1 GPU
NVIDIA RTX 4090
1445
NVIDIA L40
n/a
4 GPU
NVIDIA RTX 4090
n/a
NVIDIA L40
n/a
8 GPU
NVIDIA RTX 4090
n/a
NVIDIA L40
n/a
Redshift
1 GPU
NVIDIA RTX 4090
1.16
NVIDIA L40
n/a
4 GPU
NVIDIA RTX 4090
n/a
NVIDIA L40
n/a
8 GPU
NVIDIA RTX 4090
n/a
NVIDIA L40
n/a
Blender
1 GPU
NVIDIA RTX 4090
12123.96
NVIDIA L40
n/a
4 GPU
NVIDIA RTX 4090
n/a
NVIDIA L40
n/a
8 GPU
NVIDIA RTX 4090
n/a
NVIDIA L40
n/a
Luxmark
1 GPU
NVIDIA RTX 4090
158815
NVIDIA L40
n/a
4 GPU
NVIDIA RTX 4090
n/a
NVIDIA L40
n/a
8 GPU
NVIDIA RTX 4090
n/a
NVIDIA L40
n/a
Unreal Engine
1 GPU
NVIDIA RTX 4090
92.1
NVIDIA L40
n/a
4 GPU
NVIDIA RTX 4090
n/a
NVIDIA L40
n/a
8 GPU
NVIDIA RTX 4090
n/a
NVIDIA L40
n/a
RELION Cryo-EM Benchmarks 2024-2025 [Updated]
Total run time
1 GPU
NVIDIA RTX 4090
105.2
NVIDIA L40
n/a
4 GPU
NVIDIA RTX 4090
53.2
NVIDIA L40
n/a
8 GPU
NVIDIA RTX 4090
n/a
NVIDIA L40
n/a
BERT LARGE Training, Inference Benchmark 2024-2025 [Updated]
Training Time (FP16)
1 GPU
NVIDIA RTX 4090
n/a
NVIDIA L40
n/a
4 GPU
NVIDIA RTX 4090
n/a
NVIDIA L40
n/a
8 GPU
NVIDIA RTX 4090
n/a
NVIDIA L40
n/a
Performance (FP16)
1 GPU
NVIDIA RTX 4090
n/a
NVIDIA L40
n/a
4 GPU
NVIDIA RTX 4090
n/a
NVIDIA L40
n/a
8 GPU
NVIDIA RTX 4090
n/a
NVIDIA L40
n/a
Training Time (FP32)
1 GPU
NVIDIA RTX 4090
n/a
NVIDIA L40
n/a
4 GPU
NVIDIA RTX 4090
n/a
NVIDIA L40
n/a
8 GPU
NVIDIA RTX 4090
n/a
NVIDIA L40
n/a
Performance (FP32)
1 GPU
NVIDIA RTX 4090
n/a
NVIDIA L40
n/a
4 GPU
NVIDIA RTX 4090
n/a
NVIDIA L40
n/a
8 GPU
NVIDIA RTX 4090
n/a
NVIDIA L40
n/a
Llama3 70B Inference Benchmark 2024–2025 [Updated]
Eval rate
1 GPU
NVIDIA RTX 4090
9.95
NVIDIA L40
n/a
2 GPU
NVIDIA RTX 4090
19.99
NVIDIA L40
n/a
4 GPU
NVIDIA RTX 4090
n/a
NVIDIA L40
n/a
Technical Specifications
Board Design
| NVIDIA RTX 4090 | NVIDIA L40 | |
|---|---|---|
| Length | 336 mm | 267 mm |
| Outputs | 1x HDMI, 3x DisplayPort | 4x DisplayPort |
| Power Connectors | 1x 16-pin | 1x 16-pin |
| Slot width | Triple-slot | Dual-slot |
| TDP | 450 W | 300 W |
Clock Speeds
| NVIDIA RTX 4090 | NVIDIA L40 | |
|---|---|---|
| Boost Clock | 2520 MHz | 2490 MHz |
| GPU Clock | 2235 MHz | 735 MHz |
| Memory Clock | 21200 MHz | 18000 MHz |
Graphics Card
| NVIDIA RTX 4090 | NVIDIA L40 | |
|---|---|---|
| Bus Interface | PCIe 4.0 x16 | PCIe 4.0 x16 |
| Generation | GeForce 40 | Tesla Ada (Lxx) |
Graphics Features
| NVIDIA RTX 4090 | NVIDIA L40 | |
|---|---|---|
| DirectX | 12 Ultimate (12_2) | 12 Ultimate (12_2) |
| OpenCL | 3.0 | 3.0 |
| OpenGL | 4.6 | 4.6 |
| Shader Model | 6.7 | 6.7 |
| CUDA | 8.9 | 8.9 |
Graphics Processor
| NVIDIA RTX 4090 | NVIDIA L40 | |
|---|---|---|
| Architecture | Ada Lovelace | Ada Lovelace |
| Die Size | 608 mm2 | 608 mm2 |
| GPU Name | AD102-300-A1 | AD102-895-A1 |
| Process Size | 5 nm | 5 nm |
| Transistors | 76300 million | 76300 million |
Memory
| NVIDIA RTX 4090 | NVIDIA L40 | |
|---|---|---|
| Bandwidth | 1018 GB/s | 864 GB/s |
| Memory Bus | 384 bit | 384 bit |
| Memory Size | 24 GB | 48 GB |
| Memory Type | GDDR6X | GDDR6 |
Render Config
| NVIDIA RTX 4090 | NVIDIA L40 | |
|---|---|---|
| ROPs | 192 | 192 |
| RT Cores | 128 | 142 |
| Shading Units/ CUDA Cores | 16384 | 18176 |
| TMUs | 512 | 568 |
| Tensor Cores | 512 | 568 |
Theoretical Performance
| NVIDIA RTX 4090 | NVIDIA L40 | |
|---|---|---|
| FP16 (half) performance | 82.58 TFLOPS | 90.52 TFLOPS |
| FP32 (float) performance | 82.58 TFLOPS | 90.52 TFLOPS |
| FP64 (double) performance | 1290 GFLOPS | 1414 GFLOPS |
| Pixel Rate | 483.8 GPixel/s | 478.1 GPixel/s |
| Texture Rate | 1290 GTexel/s | 1414 GTexel/s |
Price
| NVIDIA RTX 4090 | NVIDIA L40 | |
|---|---|---|
| MSRP | 1599 $ | $ |
| Release Date | Oct 12th, 2022 | Oct 13th, 2022 |
Test bench configuration
| NVIDIA RTX 4090 | NVIDIA L40 | |
|---|---|---|
| Hardware | BIZON X5500 More details | BIZON X5500 More details |
| Software | 3D Rendering: Nvidia Driver: VRay Benchmark: Octane Benchmark: Redshift Benchmark: Blender: Luxmark: | 3D Rendering: Nvidia Driver: VRay Benchmark: Octane Benchmark: Redshift Benchmark: Blender: Luxmark: |
Recommended hardware
| NVIDIA RTX 4090 | NVIDIA L40 | |
|---|---|---|
| Best GPU workstations | BIZON X5500 | |
| Best NVIDIA GPU servers | BIZON X7000 |