RTX PRO 4000 Blackwell SFF vs Radeon 625
Contents
Primary details
GPU architecture, market segment, value for money and other general parameters compared.
| Place in the ranking | 844 | not rated |
| Place by popularity | not in top-100 | not in top-100 |
| Power efficiency | 3.92 | no data |
| Architecture | GCN 3.0 (2014−2019) | Blackwell 2.0 (2025) |
| GPU code name | Polaris 24 | GB203 |
| Market segment | Laptop | Workstation |
| Release date | 13 May 2019 (6 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 | 384 | 8960 |
| Core clock speed | 730 MHz | 790 MHz |
| Boost clock speed | 1024 MHz | 1337 MHz |
| Number of transistors | 1,550 million | 45,600 million |
| Manufacturing process technology | 28 nm | 5 nm |
| Power consumption (TDP) | 50 Watt | 70 Watt |
| Texture fill rate | 24.58 | 374.4 |
| Floating-point processing power | 0.7864 TFLOPS | 23.96 TFLOPS |
| ROPs | 8 | 96 |
| TMUs | 24 | 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 3.0 x8 | PCIe 5.0 x8 |
| Length | no data | 167 mm |
| Width | no data | 1-slot |
| Supplementary power connectors | None | 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 | DDR3 | GDDR7 |
| Maximum RAM amount | 2 GB | 24 GB |
| Memory bus width | 64 Bit | 192 Bit |
| Memory clock speed | 900 MHz | 1125 MHz |
| Memory bandwidth | 14.4 GB/s | 432.0 GB/s |
| Shared memory | - | - |
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 | 12 (12_0) | 12 Ultimate (12_2) |
| Shader Model | 6.3 | 6.8 |
| OpenGL | 4.6 | 4.6 |
| OpenCL | 2.0 | 3.0 |
| Vulkan | 1.2.131 | 1.4 |
| CUDA | - | 12.0 |
| DLSS | - | + |
Pros & cons summary
| Recency | 13 May 2019 | 11 August 2025 |
| Maximum RAM amount | 2 GB | 24 GB |
| Chip lithography | 28 nm | 5 nm |
| Power consumption (TDP) | 50 Watt | 70 Watt |
Radeon 625 has 40% lower power consumption.
RTX PRO 4000 Blackwell SFF, on the other hand, has an age advantage of 6 years, a 1100% higher maximum VRAM amount, and a 460% more advanced lithography process.
We couldn't decide between Radeon 625 and RTX PRO 4000 Blackwell SFF. We've got no test results to judge.
Be aware that Radeon 625 is a notebook graphics card while RTX PRO 4000 Blackwell SFF 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.
