Radeon Instinct MI455X vs GeForce RTX 5090 D
Contents
Primary details
GPU architecture, market segment, value for money and other general parameters compared.
| Place in the ranking | 1 | not rated |
| Place by popularity | not in top-100 | not in top-100 |
| Cost-effectiveness evaluation | 22.47 | no data |
| Power efficiency | 13.43 | no data |
| Architecture | Blackwell 2.0 (2025−2026) | CDNA 5.0 (2026) |
| GPU code name | GB202 | MI450 256CU |
| Market segment | Desktop | Workstation |
| Release date | 30 January 2025 (1 year ago) | 23 July 2026 (recently) |
| Launch price (MSRP) | $2,299 | no data |
Cost-effectiveness evaluation
The higher the ratio, the better. We use the manufacturer's recommended prices.
Performance to price scatter graph
Currently popular graphics cards are shown for comparison.
Detailed specifications
General parameters such as number of shaders, GPU core base clock and boost clock speeds, manufacturing process, texturing and calculation speed. Real power consumption of some graphics cards can well exceed their nominal TDP, especially if overclocked.
| Pipelines / CUDA cores | 21760 | 32768 |
| Core clock speed | 2017 MHz | 1000 MHz |
| Boost clock speed | 2407 MHz | 2400 MHz |
| Number of transistors | 92,200 million | 320,000 million |
| Manufacturing process technology | 5 nm | 2 nm |
| Power consumption (TDP) | 575 Watt | 1400 Watt |
| Texture fill rate | 1,636.8 | 2,457.6 |
| Floating-point processing power | 104.8 TFLOPS | 157.3 TFLOPS |
| ROPs | 176 | no data |
| TMUs | 680 | 1024 |
| Tensor Cores | 680 | no data |
| Ray Tracing Cores | 170 | no data |
| L1 Cache | 21.3 MB | 8 MB |
| L2 Cache | 96 MB | 192 MB |
Form factor & compatibility
Information on compatibility with other computer components. Useful when choosing a future computer configuration or upgrading an existing one. For desktop graphics cards it's interface and bus (motherboard compatibility), additional power connectors (power supply compatibility).
| Interface | PCIe 5.0 x16 | PCIe 5.0 x16 |
| Length | 304 mm | no data |
| Width | 2-slot | EAM Module |
| Supplementary power connectors | 1x 16-pin | None |
VRAM capacity and type
Parameters of VRAM installed: its type, size, bus, clock and resulting bandwidth. Integrated GPUs have no dedicated video RAM and use a shared part of system RAM.
| Memory type | GDDR7 | HBM4 |
| Maximum RAM amount | 32 GB | 432 GB |
| Memory bus width | 512 Bit | 24576 Bit |
| Memory clock speed | 1750 MHz | 1900 MHz |
| Memory bandwidth | 1.79 TB/s | 23.3 TB/s |
| Resizable BAR | + | - |
Connectivity and outputs
This section shows the types and number of video connectors on each GPU. The data applies specifically to desktop reference models (for example, NVIDIA’s Founders Edition). OEM partners often modify both the number and types of ports. On notebook GPUs, video‐output options are determined by the laptop’s design rather than the graphics chip itself.
| Display Connectors | 1x HDMI 2.1b, 3x DisplayPort 2.1b | No outputs |
| HDMI | + | - |
API and SDK support
List of supported 3D and general-purpose computing APIs, including their specific versions.
| DirectX | 12 Ultimate (12_2) | N/A |
| Shader Model | 6.8 | N/A |
| OpenGL | 4.6 | N/A |
| OpenCL | 3.0 | 3.0 |
| Vulkan | 1.4 | N/A |
| CUDA | 12.0 | - |
| DLSS | + | - |
Pros & cons summary
| Recency | 30 January 2025 | 23 July 2026 |
| Maximum RAM amount | 32 GB | 432 GB |
| Chip lithography | 5 nm | 2 nm |
| Power consumption (TDP) | 575 Watt | 1400 Watt |
RTX 5090 D has 143% lower power consumption.
Instinct MI455X, on the other hand, has an age advantage of 1 year, a 1250% higher maximum VRAM amount, and a 150% more advanced lithography process.
We couldn't decide between GeForce RTX 5090 D and Radeon Instinct MI455X. We've got no test results to judge.
Be aware that GeForce RTX 5090 D is a desktop graphics card while Radeon Instinct MI455X is a workstation one.
Other comparisons
We selected several comparisons of graphics cards with performance close to those reviewed, providing you with more options to consider.
