RTX PRO 4000 Blackwell SFF vs Radeon Instinct MI100
Contents
Primary details
GPU architecture, market segment, value for money and other general parameters compared.
| Place in the ranking | not rated | not rated |
| Place by popularity | not in top-100 | not in top-100 |
| Architecture | CDNA 1.0 (2020) | Blackwell 2.0 (2025) |
| GPU code name | Arcturus | GB203 |
| Market segment | Workstation | Workstation |
| Release date | 16 November 2020 (4 years ago) | 11 August 2025 (recently) |
Detailed specifications
General parameters such as number of shaders, GPU core base clock and boost clock speeds, manufacturing process, texturing and calculation speed. Note that power consumption of some graphics cards can well exceed their nominal TDP, especially when overclocked.
| Pipelines / CUDA cores | 7680 | 8960 |
| Core clock speed | 1000 MHz | 790 MHz |
| Boost clock speed | 1502 MHz | 1337 MHz |
| Number of transistors | 50,000 million | 45,600 million |
| Manufacturing process technology | 7 nm | 5 nm |
| Power consumption (TDP) | 300 Watt | 70 Watt |
| Texture fill rate | 721.0 | 374.4 |
| Floating-point processing power | 23.07 TFLOPS | 23.96 TFLOPS |
| ROPs | 64 | 96 |
| TMUs | 480 | 280 |
| Tensor Cores | no data | 280 |
| Ray Tracing Cores | no data | 70 |
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 4.0 x16 | PCIe 5.0 x8 |
| Length | 267 mm | 167 mm |
| Width | 2-slot | 1-slot |
| Supplementary power connectors | 2x 8-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 | HBM2 | GDDR7 |
| Maximum RAM amount | 32 GB | 24 GB |
| Memory bus width | 4096 Bit | 192 Bit |
| Memory clock speed | 1200 MHz | 1125 MHz |
| Memory bandwidth | 1,229 GB/s | 432.0 GB/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 | No outputs | 4x mini-DisplayPort 2.1b |
API and SDK compatibility
List of supported 3D and general-purpose computing APIs, including their specific versions.
| DirectX | N/A | 12 Ultimate (12_2) |
| Shader Model | N/A | 6.8 |
| OpenGL | N/A | 4.6 |
| OpenCL | 2.1 | 3.0 |
| Vulkan | N/A | 1.4 |
| CUDA | - | 12.0 |
| DLSS | - | + |
Pros & cons summary
| Recency | 16 November 2020 | 11 August 2025 |
| Maximum RAM amount | 32 GB | 24 GB |
| Chip lithography | 7 nm | 5 nm |
| Power consumption (TDP) | 300 Watt | 70 Watt |
Instinct MI100 has a 33.3% higher maximum VRAM amount.
RTX PRO 4000 Blackwell SFF, on the other hand, has an age advantage of 4 years, a 40% more advanced lithography process, and 328.6% lower power consumption.
We couldn't decide between Radeon Instinct MI100 and RTX PRO 4000 Blackwell SFF. We've got no test results to judge.
Other comparisons
We selected several comparisons of graphics cards with performance close to those reviewed, providing you with more options to consider.
