AMD Radeon RX 9070 GRE vs GeForce RTX 5070
VS
AMD Radeon RX 9070 GRE
Navi 48
Nvidia GeForce RTX 5070
GB205-300
We compared two discrete desktop gaming GPUs: the Radeon RX 9070 GRE with 48 pipelines and 3072 shaders against the 8 months older GeForce RTX 5070 that utilizes 48 pipelines and 6144 shaders. Here you will find complete details about specs, efficiency, performance tests, and more.
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
Scores marked with a red asterisk (**) represent initial estimates from early testing.
Value for money (Beta)
Key differences
Reasons to consider the GeForce RTX 5070
- Supports Nvidia DLSS 4 technology
- Features 96 more tensor cores for effective ML and AI workloads
- Has 56% higher memory bandwidth: 672 vs 432 GB/s
- Has 2x more shading units (6144 vs 3072)
Gaming Performance
Frame rate comparison across popular AAA titles at different resolutionsGames
| Forza Horizon 5 | - | 197 140 130 97 |
| The Witcher 3 | - | 216 183 131 71 |
| Counter-Strike 2 | - | 293 210 160 81 |
| Far Cry 6 | - | 164 145 128 82 |
| Hogwarts Legacy | - | 140 117 92 51 |
| CoD: Modern Warfare III | - | 178 171 136 89 |
| Ghost of Tsushima | - | 123 107 89 58 |
| Cyberpunk 2077 | - | 164 144 99 43 |
| Shadow of the Tomb Raider | - | 253 238 189 98 |
| 1080p High | - | 192 |
| 1080p Ultra | - | 162 |
| 1440p Ultra | - | 128 |
| 4K Ultra | - | 74 |
| Margin of Error | High | Low |
Benchmarks
Graphics cards’ performance in recent benchmarking apps3D Mark
Steel Nomad Lite Score
| Time Spy | - | 22708 |
| Solar Bay | - | 107004 |
| Port Royal | - | 14498 |
| Fire Strike | - | 58324 |
| Wild Life Extreme | - | 43879 |
| Night Raid | - | 172064 |
GeekBench 6 GPU Compute
GB6 Compute Score
| Background Blur | - | 254.4 img/sec |
| Face Detection | - | 180.6 img/sec |
| Horizon Detection | - | 7.91 Gpixels/sec |
| Edge Detection | - | 10.2 Gpixels/sec |
| Gaussian Blur | - | 10.7 Gpixels/sec |
| Feature Matching | - | 1.84 Gpixels/sec |
| Stereo Matching | - | 975.7 Gpixels/sec |
| Particle Physics | - | 27816.5 FPS |
| API | - | OpenCL |
Passmark Graphics
G3D Mark Score
| G2D Mark | 1261 | 1289 |
| DirectX 11 | 254 FPS | 279 FPS |
| DirectX 12 | 56 FPS | 106 FPS |
| GPU Compute | 14796 Ops/s | 14323 Ops/s |
Blender
Blender GPU
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) | - | 12270 |
| Image Segmentation (HP) | - | 30237 |
| Image Super Resolution (Q) | - | 35585 |
| Face Detection (HP) | - | 62662 |
| Pose Estimation (Q) | - | 145190 |
| Text Classification (SP) | - | 4198 |
| Machine Translation (HP) | - | 6233 |
| Object Detection (SP) | - | 16709 |
| Depth Estimation (Q) | - | 58300 |
| Style Transfer (SP) | - | 246598 |
| Framework | - | ONNX |
| Backend | - | DirectML |
Specifications
Technical specifications of Radeon RX 9070 GRE and GeForce RTX 5070General
| Vendor | Amd | Nvidia |
| Build | Discrete | Discrete |
| Released | August 27, 2025 | January 7, 2025 |
| Launch price (MSRP) | $549 | $549 |
| Case | Desktop | Desktop |
| Purpose | Gaming | Gaming |
| Segment | High-end | Mid-range |
| Architecture | RDNA 4.0 | Blackwell 2.0 |
| GPU Codename | Navi 48 | GB205-300 |
| Rival Equivalent | - GeForce RTX 5060 Ti (8GB) | - Radeon RX 9070 |
| Recommended CPU | - Intel Core Ultra 5 245K or above | - Intel Core Ultra 5 245K or above |
Desktop GPU rating (25th and 21st place)
Graphics Processing Unit
| Base Clock | 1420 MHz | 2160 MHz |
| Boost Clock | 2790 MHz | 2512 MHz |
| Shading Units | 3072 | 6144 |
| Texture Mapping Units (TMUs) | 192 | 192 |
| Render Output Units (ROPs) | 96 | 80 |
| Compute Units (Pipelines) | 48 | 48 |
| Tensor Cores | 96 | 192 |
| Ray-tracing Cores | 48 | 48 |
| L1 Cache | 128KB per cluster | 128KB per cluster |
| L2 Cache | 8MB shared | 48MB shared |
| Instructions Per Cycle | 4 IPC | 2 IPC |
Raw Performance
| Pixel Fill Rate | 268 GPixel/s | 201 GPixel/s |
| Texture Fill Rate | 536 GTexel/s | 482 GTexel/s |
FLOPS (FP32)
Differences in GPUs' instructions per clock make TFLOPS values not directly comparable.
Physical
| Interface | PCIe 5.0 x16 | PCIe 5.0 x16 |
| TGP | 220 W | 250 W |
| Manufacturing | TSMC | TSMC |
| Fabrication Process | 4 nm | 4 nm |
| Die Size | 357 mm² | 263 mm² |
| Transistor Count | 53.9 billion | 31.1 billion |
| Transistor Density | 150.98 MTr/mm² | 118.25 MTr/mm² |
Memory
| Memory Type | GDDR6 | GDDR7 |
| Memory Size | 12 GB | 12 GB |
| Memory Clock | 2250 MHz | 1750 MHz |
| Effective Memory Speed | 18000 Mbps | 28000 Mbps |
| Bus | 192-bit | 192-bit |
| ECC | No | No |
Memory Bandwidth
API
| DirectX | 12 | 12.2 |
| Vulkan | 1.4 | 1.4 |
| OpenGL | 4.6 | 4.6 |
| OpenCL | 2.2 | 3.0 |
| CUDA | No | 12.0 |
| Ray Tracing | Yes | Yes |
| DLSS | No | DLSS 4 |
| DisplayPort | 2.1a | 2.1b |
Cast your vote
Total votes: < 1