Quick/Comparisons
A budget Nvidia GeForce MX250 graphics card with limited performance overall. Suited for casual gaming and basic graphics tasks.
| Gaming | 5 /100 |
| Workstation | 9 /100 |
| AI/ML | 3 /100 |
| Energy Efficiency | 68 |
Looking for more comparisons?
Compare GeForce MX250 →| Quick Comparison Final Score | 16 /100 |
| Graphics Processor | GP108 |
| Cores | 384 |
| TMUs / ROPs | 24 / 16 |
| Blender GPU | 87.25 |
| Forza Horizon 5 | - |
| The Witcher 3 | - |
| Counter-Strike 2 | 17 |
| Far Cry 6 | 16 |
| Hogwarts Legacy | - |
| Call of Duty | 23 |
| Ghost of Tsushima | - |
| Cyberpunk 2077 | — |
| Tomb Raider | - |
| Average FPS | 10 |
| 1080p High | 14 |
| 1080p Ultra | 10 |
| 1440p Ultra | - |
| 4K Ultra | - |
| Margin of Error | Medium |
| GB6 Compute Score | 10870 |
| Background Blur | 27.8 img/sec |
| Face Detection | 13 img/sec |
| Horizon Detection | 0.4 Gpixels/sec |
| Edge Detection | 0.59 Gpixels/sec |
| Gaussian Blur | 0.64 Gpixels/sec |
| Feature Matching | 0.12 Gpixels/sec |
| Stereo Matching | 34 Gpixels/sec |
| Particle Physics | 960.9 FPS |
| API | OpenCL |
| G3D Mark Score | 2376 |
| G2D Mark | 238 |
| DirectX 11 | 18 FPS |
| DirectX 12 | 12 FPS |
| GPU Compute | 1011 Ops/s |
| GB6 ML Single Precision | 1840 |
| GB6 ML Half Precision | 1885 |
| GB6 ML Quantized | 1401 |
| Image Classification (SP) | 405 |
| Image Segmentation (HP) | 968 |
| Image Super Resolution (Q) | 2273 |
| Face Detection (HP) | 1652 |
| Pose Estimation (Q) | 8039 |
| Text Classification (SP) | 837 |
| Machine Translation (HP) | 944 |
| Object Detection (SP) | 552 |
| Depth Estimation (Q) | 3111 |
| Style Transfer (SP) | 17477 |
| Framework | ONNX |
| Backend | DirectML |
| Pixel Fill Rate | 25 GPixel/s |
| Texture Fill Rate | 38 GTexel/s |
| FLOPS (FP32) | 1.2 TFLOPS |
| Memory Type | GDDR5 |
| Memory Size | 2 GB |
| Memory Clock | 1502 MHz |
| Effective Memory Speed | 6000 Mbps |
| Bus | 64-bit |
| ECC | — |
| Memory Bandwidth | 48 GB/s |
| Interface | PCIe 3.0 x4 |
| TGP | 25 W |
| Manufacturing | Samsung |
| Fabrication Process | 14 nm |
| Die Size | 74 mm² |
| Transistor Count | 1.8 billion |
| Transistor Density | 24.32 MTr/mm² |
| DirectX | 12 |
| Vulkan | 1.2 |
| OpenGL | 4.6 |
| OpenCL | 1.2 |
| CUDA | 6.1 |
| Ray Tracing | No |
| DLSS | No |
| DisplayPort | 1.4 |
| Vendor | Nvidia |
| Build | Discrete |
| Released | February 21, 2009 |
| Case | Laptop |
| Purpose | Gaming |
| Segment | Entry-level |
| Architecture | Pascal |
| GPU Codename | GP108 |
| Recommended CPU | -Intel Core i7 1065G7or above |
| Steel Nomad Lite Score | 896 |
| Time Spy | 1078 |
| Solar Bay | — |
| Port Royal | — |
| Fire Strike | 3619 |
| Wild Life Extreme | 2227 |
| Night Raid | 14485 |
| Base Clock | 1519 MHz |
| Boost Clock | 1582 MHz |
| Shading Units | 384 |
| Texture Mapping Units (TMUs) | 24 |
| Render Output Units (ROPs) | 16 |
| Compute Units (Pipelines) | 3 |
| Tensor Cores | No |
| Ray-tracing Cores | No |
| L1 Cache | 48KB per cluster |
| L2 Cache | 512KB shared |
| Instructions Per Cycle | 2 IPC |
84/100