Nvidia RTX Pro 3000 Laptop (Blackwell) vs GeForce RTX 4070 Laptop
VS
Nvidia RTX Pro 3000 Laptop (Blackwell)
GB205-300-A1
Nvidia GeForce RTX 4070 Laptop
AD106 (GN21-X6)
We compared two discrete laptop graphics cards: the Nvidia RTX Pro 3000 Mobile (Blackwell) 12 GB with 46 pipelines and 5888 shaders against the 2 years and 1 month older GeForce RTX 4070 Mobile 8 GB that utilizes 36 pipelines and 4608 shaders. Here you will find complete details about specs, efficiency, performance tests, and more.
Some GPUs come in different configurations. For greater precision, pick a specific config below
Review
Gaming
Performance in DirectX, OpenCL, and Vulkan games
Workstation
Perf. in 3D modeling, video editing and rendering apps
AI/ML
Capabilities for machine learning and AI-related tasks
Energy Efficiency
Power consumption efficiency in different scenarios
NanoReview Final Score
Overall video card score
Key differences
Reasons to consider the Nvidia RTX Pro 3000 Laptop (Blackwell)
- 43% higher maximum theoretical performance (28.6 vs 20 TFLOPS)
- Includes 4 GB more video memory
- Has 2.6x higher memory bandwidth: 672 vs 256 GB/s
- Achieves 46% more points in the GeekBench 6 Compute test (167K vs 115K)
- Supports Nvidia DLSS 4 technology
- Has 28% more shading units (5888 vs 4608)
Gaming Performance
Frame rate comparison across popular AAA titles at different resolutionsGames
| Forza Horizon 5 | 137 87 76 56 |
139 89 77 56 |
| The Witcher 3 | 154 130 93 50 |
164 129 95 53 |
| Counter-Strike 2 | 224 173 126 67 |
236 183 130 71 |
| Far Cry 6 | 121 107 88 49 |
121 109 87 51 |
| Hogwarts Legacy | 88 73 56 29 |
91 77 60 28 |
| CoD: Modern Warfare III | 111 104 74 46 |
111 105 77 51 |
| Ghost of Tsushima | 85 70 50 30 |
85 69 53 29 |
| Cyberpunk 2077 | 96 86 51 24 |
99 84 53 25 |
| Shadow of the Tomb Raider | 146 133 100 51 |
147 137 100 53 |
| 1080p High | 129 | 133 |
| 1080p Ultra | 107 | 109 |
| 1440p Ultra | 79 | 81 |
| 4K Ultra | 45 | 46 |
| Margin of Error | High | Low |
Benchmarks
Graphics cards’ performance in recent benchmarking apps3D Mark
Steel Nomad Lite Score
| Time Spy | 12518 | 12514 |
| Solar Bay | 66557 | 54815 |
| Port Royal | 10955 | 7416 |
| Fire Strike | 32060 | 30057 |
| Wild Life Extreme | - | 23262 |
| Night Raid | - | 113672 |
GeekBench 6 OpenCL
GB6 Compute Score
| Background Blur | 238.8 img/sec | 158.5 img/sec |
| Face Detection | 147.6 img/sec | 116.5 img/sec |
| Horizon Detection | 6.73 Gpixels/sec | 4.35 Gpixels/sec |
| Edge Detection | 9.27 Gpixels/sec | 5.67 Gpixels/sec |
| Gaussian Blur | 10.7 Gpixels/sec | 6.44 Gpixels/sec |
| Feature Matching | 1.68 Gpixels/sec | 1.36 Gpixels/sec |
| Stereo Matching | 778.4 Gpixels/sec | 528.5 Gpixels/sec |
| Particle Physics | 23023.5 FPS | 16372.3 FPS |
| API | OpenCL | OpenCL |
Cinebench 2024 GPU
Cinebench 2024 GPU
Passmark Graphics
G3D Mark Score
| G2D Mark | 826 | 766 |
| DirectX 11 | 212 FPS | 177 FPS |
| DirectX 12 | 94 FPS | 85 FPS |
| GPU Compute | 7266 Ops/s | 8365 Ops/s |
Blender
Blender GPU
Recent User Tests
Nvidia RTX Pro 3000 Laptop (Blackwell)
No benchmark results yet
GeForce RTX 4070 Laptop
| Date | Benchmark | Result |
|---|---|---|
| 📘 2026-06-05 (Wahid) | Geekbench 6 OpenCL | 122515 |
| 📘 2026-06-05 (Wahid) | Geekbench 6 OpenCL | 35869 |
Artificial Intelligence Tests
Performance in machine learning and artificial intelligence tasksGeekBench 6 ML
GB6 ML Single Precision
GB6 ML Half Precision
GB6 ML Quantized
| Image Classification (SP) | 9790 | 8642 |
| Image Segmentation (HP) | 28672 | 20635 |
| Image Super Resolution (Q) | 26553 | 26378 |
| Face Detection (HP) | 53161 | 39776 |
| Pose Estimation (Q) | 120306 | 105983 |
| Text Classification (SP) | 3406 | 2995 |
| Machine Translation (HP) | 5910 | 4507 |
| Object Detection (SP) | 12573 | 10198 |
| Depth Estimation (Q) | 47957 | 41078 |
| Style Transfer (SP) | 199379 | 190315 |
| Framework | ONNX | ONNX |
| Backend | DirectML | DirectML |
Specifications
Technical specifications of Nvidia RTX Pro 3000 Laptop (Blackwell) and GeForce RTX 4070 LaptopGeneral
| Vendor | Nvidia | Nvidia |
| Build | Discrete | Discrete |
| Released | March 18, 2025 | February 22, 2023 |
| Case | Laptop | Laptop |
| Purpose | Professional | Gaming |
| Segment | Mid-range | High-end |
| Architecture | Blackwell 2.0 | Ada Lovelace |
| GPU Codename | GB205-300-A1 | AD106 (GN21-X6) |
| Successor | - | - GeForce RTX 5070 Laptop |
| Recommended CPU | - Intel Core Ultra 7 265HX or above | - Intel Core Ultra 7 155H or above |
Laptop GPU ranking (13th and 23rd place)
Graphics Processing Unit
| Base Clock | 848-1605 MHz (configurable) | 735-2070 MHz (configurable) |
| Boost Clock | 1605-2430 MHz (configurable) | 1230-2175 MHz (configurable) |
| Shading Units | 5888 | 4608 |
| Texture Mapping Units (TMUs) | 184 | 144 |
| Render Output Units (ROPs) | 80 | 48 |
| Compute Units (Pipelines) | 46 | 36 |
| Tensor Cores | 184 | 144 |
| Ray-tracing Cores | 46 | 36 |
| L1 Cache | 128KB per cluster | 128KB per cluster |
| L2 Cache | 48MB shared | 32MB shared |
| Instructions Per Cycle | 2 IPC | 2 IPC |
Raw Performance
| Pixel Fill Rate | 128-194 GPixel/s | 59-104 GPixel/s |
| Texture Fill Rate | 295-447 GTexel/s | 177-313 GTexel/s |
FLOPS (FP32)
Physical
| Interface | PCIe 5.0 x16 | PCIe 4.0 x16 |
| TGP | 60-140 W (configurable) | 35-140 W (configurable) |
| Manufacturing | TSMC | TSMC |
| Fabrication Process | 5 nm | 5 nm |
| Die Size | 263 mm² | 186 mm² |
| Transistor Count | 31.1 billion | 22.9 billion |
| Transistor Density | 118.25 MTr/mm² | 123.12 MTr/mm² |
Memory
| Memory Type | GDDR7 | GDDR6 |
| Memory Size | 12 GB | 8 GB |
| Memory Clock | 1750 MHz | 2000 MHz |
| Effective Memory Speed | 28000 Mbps | 16000 Mbps |
| Bus | 192-bit | 128-bit |
| ECC | No | No |
Memory Bandwidth
API
| DirectX | 12.2 | 12 |
| Vulkan | 1.4 | 1.3 |
| OpenGL | 4.6 | 4.6 |
| OpenCL | 3.0 | 3.0 |
| CUDA | 12 | 8.9 |
| Ray Tracing | Yes | Yes |
| DLSS | DLSS 4 | DLSS 3 |
| DisplayPort | 2.1b | 1.4a |
Cast your vote
Total votes: < 1