GeForce RTX 4090 VS GeForce RTX 5090
We are comparing the GeForce RTX 4090 (NVIDIA) with the GeForce RTX 5090 (NVIDIA). Scroll down to the pros and cons of each graphics card, in-depth specs, and gaming benchmarks.
Last Updated: September 4, 2026
Overall Comparison
KEY DIFFERENCES OF GeForce RTX 4090 AND GeForce RTX 5090
Here’s how GeForce RTX 4090 and GeForce RTX 5090 stack up against each other in their most impactful areas.
Reasons to consider the GeForce RTX 4090
-
5% higher Boost Clock (2520 MHz vs 2407 MHz)
-
6% higher PassMark GPU Compute score (26273 vs 24875)
Reasons to consider the GeForce RTX 5090
-
19% higher 4K Ultra FPS (125 vs 105)
-
18% higher 1440p Ultra FPS (207 vs 176)
-
16% higher 1080p Ultra FPS (253 vs 218)
-
29% higher 3DMark Time Spy score (46966 vs 36315)
-
45% higher 3DMark Port Royal score (37963 vs 26128)
SPECIFICATION TABLE COMPARISON OF GeForce RTX 4090 AND GeForce RTX 5090
Below is the detailed specification table comparing GeForce RTX 4090 and GeForce RTX 5090.
Note: Highlighted specs mean that GPU wins that spec row.
If neither side is highlighted, it is a tie or there is not enough data for both specs.
| Specification | GeForce RTX 4090 | GeForce RTX 5090 |
|---|---|---|
|
General Info
|
||
| GPU Name: | GeForce RTX 4090 | GeForce RTX 5090 |
| GPU Type: | Discrete | Discrete |
| Codename: | AD102 | GB202-300 |
| Vendor: | NVIDIA | NVIDIA |
| Foundry: | TSMC | TSMC |
| Architecture: | Ada Lovelace | Blackwell 2.0 |
| Generation: | GeForce 40 | GeForce 50 |
| Release Date: | September 20, 2022 | January 30, 2025 |
|
Board Design
|
||
| Bus Interface: | PCIe 4.0 x16 | PCIe 5.0 x16 |
| PCIe Version: | 4.0 | 5.0 |
| PCIe Lanes: | x16 | x16 |
| Process Size: | 5 nm | 5 nm |
| Transistors: | 76.3 billion | 92.2 billion |
| Density: | 125.29 MTr/mm² | 122.93 MTr/mm² |
| Die Size: | 609 | 750 |
| Board Length: | 304 mm | 304 mm |
| Board Width: | 61 mm | 40 mm |
| Board Height: | 137 mm | 137 mm |
| Total Graphics Power (TGP): | 450 W | 575 W |
|
Graphics Processor
|
||
| Compute Units: | 128 | 170 |
| Shading Units: | 16384 | 21760 |
| Texture Mapping Units (TMUs): | 512 | 680 |
| Render Output Units (ROPs): | 176 | 176 |
| Tensor Cores: | 512 | 680 |
| Ray Tracing Cores: | 128 | 170 |
| Base Clock: | 2235 | 2017 |
| Boost Clock: | 2520 | 2407 |
| L1 Cache: | 128KB per cluster | 128KB per cluster |
| L2 Cache: | 72MB shared | 96MB shared |
|
Raw Performance
|
||
| Pixel Fill Rate: | 443.5 GPixel/s | 423.6 GPixel/s |
| Texture Fill Rate: | 1,290.2 GTexel/s | 1,636.8 GTexel/s |
| FP32 Performance: | 82.58 TFLOPs | 104.8 TFLOPs |
|
Memory
|
||
| Memory Type: | GDDR6X | GDDR7 |
| Memory Size: | 24 GB | 32 GB |
| Memory Bus Width: | 384-bit | 512-bit |
| Memory Clock: | 2625 | 1750 |
| Effective Memory Speed: | 21000 | 28000 |
| Bandwidth: | 1010 | 1792 |
|
APIs
|
||
| DirectX: | 12 | 12.2 |
| Vulkan: | 1.3 | 1.4 |
| OpenGL: | 4.6 | 4.6 |
| OpenCL: | 3.0 | 3.0 |
| CUDA: | 8.9 | 12.0 |
| Ray Tracing: | Yes | Yes |
| DLSS: | DLSS 3 | DLSS 4 |
| DisplayPort: | 1.4a | 2.1b |
|
Gaming FPS
|
||
| 1080p Ultra (FPS): | 218 | 253 |
| 1440p Ultra (FPS): | 176 | 207 |
| 4K Ultra (FPS): | 105 | 125 |
|
Benchmarks
|
||
|
3D Mark
|
||
| Steel Nomad Lite | 42154 | 52760 |
| Time Spy | 36315 | 46966 |
| Solar Bay | 187290 | 234280 |
| Port Royal | 26128 | 37963 |
| Fire Strike | 72272 | 87056 |
| Wild Life Extreme | 85225 | 108665 |
| Night Raid | 195744 | 207642 |
|
GeekBench 6 OpenCL
|
||
| Compute Score | 316301 | 419309 |
| Background Blur | 318.9 img/sec | 333 img/sec |
| Face Detection | 222.8 img/sec | 298.5 img/sec |
| Horizon Detection | 14.2 Gpixels/sec | 21.7 Gpixels/sec |
| Edge Detection | 21 Gpixels/sec | 33.7 Gpixels/sec |
| Gaussian Blur | 23.9 Gpixels/sec | 34.5 Gpixels/sec |
| Feature Matching | 2.38 Gpixels/sec | 2.53 Gpixels/sec |
| Stereo Matching | 1950 Gpixels/sec | 2850 Gpixels/sec |
| Particle Physics | 48090.4 FPS | 59146.9 FPS |
|
Passmark Graphics
|
||
| G3D Mark | 38098 | 38868 |
| G2D Mark | 1303 | 1412 |
| DirectX 11 | 324 | 341 |
| DirectX 12 | 151 | 177 |
| GPU Compute | 26273 Ops/s | 24875 Ops/s |
|
GFXBench 5
|
||
| Aztec Ruins High (4K) | 615.90 | 624.60 |
| Aztec Ruins Normal | 1617.60 | 1568.70 |
| Car Chase | 1228.60 | 1292.00 |
| Manhattan 3.1.1 (1440p) | 1234.60 | 1244.60 |
| T-Rex | 6286.00 | 5493.70 |
| Tessellation | 12119.10 | 13760.70 |
| ALU 2 | 15073.40 | 15791.90 |
| Driver Overhead 2 | 303.70 | 332.10 |
| API | OpenGL | OpenGL |
| OS | Windows | Windows |
|
Blender (Rendering)
|
||
| Blender Score | 11795 | 15030 |
|
GeekBench 6 ML
|
||
| Single Precision | 42962 | 53861 |
| Half Precision | 59636 | 81080 |
| Quantized | 31670 | 38938 |
| Image Classification | 15600 | 18948 |
| Image Segmentation | 36657 | 55109 |
| Image Super Resolution | 48114 | 55109 |
| Face Detection | 78111 | 114685 |
| Pose Estimation | 289639 | 344729 |
| Text Classification | 4074 | 4464 |
| Machine Translation | 6349 | 8184 |
| Object Detection | 21186 | 27802 |
| Depth Estimation | 75599 | 85572 |
| Style Transfer | 601428 | 687080 |
| Framework | ONNX | ONNX |
| Backend | DirectML | DirectML |
|
FurMark
|
||
| 1080p Vulkan | 27475 | 34140 |
| 1440p Vulkan | 20403 | 25722 |
| 2160p Vulkan | 11883 | 14577 |
| 1080p OpenGL | 24851 | 29565 |
| 1440p OpenGL | 19078 | 23398 |
| 2160p OpenGL | 11425 | 14161 |
Tip: ▲ indicates a higher-is-better spec advantage, while ▼ indicates a lower-is-better advantage.
PERFORMANCE CHART COMPARISON OF GeForce RTX 4090 AND GeForce RTX 5090
The charts below visualize how GeForce RTX 4090 and GeForce RTX 5090 stack up in key performance metrics.
Radar Chart Comparison
Bullet Chart Comparison
Release Timeline of GeForce RTX 4090 and GeForce RTX 5090
GeForce RTX 4090 was released 2 years, 4 months, 10 days earlier than GeForce RTX 5090.
Value For Money Calculator
Prices use available launch or current data. Enter your own prices to see which GPU gives better value.
GeForce RTX 4090
GeForce RTX 5090
Calculating...
Domain Score Summary
Scores are generated by Comparigon's scoring algorithm and are intended as a high-level guide. They may not map 1-to-1 with the individual spec-by-spec comparisons shown in the table.
Our Verdict
In conclusion, the GeForce RTX 5090 is clearly better overall compared to the GeForce RTX 4090. It comes out ahead across every comparable key area.
Try to Compare Again One of These GPUs With Another GPU
Pick which GPU to keep, then search for the other GPU to compare.
Disclaimer: Specs, pricing, and benchmarks are sourced from official manufacturer data and verified test results. While we strive for precision, product details and pricing may change. Comparigon may earn an affiliate commission on qualifying purchases.