Apple M4 Ultra GPU (80-core) vs Nvidia RTX Pro 6000 Blackwell
[ VERDICT ]
Winner: Nvidia RTX Pro 6000 Blackwell (98/100)
Both devices are evenly matched with an overall score of 98/100.
Benchmark Performance Comparison
Normalized Lab Scores| Benchmark | Apple M4 Ultra GPU (80-core) | Nvidia RTX Pro 6000 Blackwell |
|---|---|---|
| Efficiency | — | 49.0 Poor |
| Gaming | — | 99.0 Excellent |
Detailed Technical Specifications
API
| Specification | Apple M4 Ultra GPU (80-core) | Nvidia RTX Pro 6000 Blackwell |
|---|---|---|
| DLSS | No | DLSS 4 |
| Ray Tracing | Yes | Yes |
| CUDA | — | 12 |
| DirectX | — | 12.2 |
| DisplayPort | — | 2.1b |
| OpenCL | — | 3.0 |
| OpenGL | — | 4.6 |
| Vulkan | — | 1.4 |
GeekBench 6 OpenCL
| Specification | Apple M4 Ultra GPU (80-core) | Nvidia RTX Pro 6000 Blackwell |
|---|---|---|
| API | OpenCL | OpenCL |
| Background Blur | — | 366.1 img/sec |
| Edge Detection | — | 35.2 Gpixels/sec |
| Face Detection | — | 248 img/sec |
| Feature Matching | — | 2.47 Gpixels/sec |
| Gaussian Blur | — | 45 Gpixels/sec |
| Horizon Detection | — | 23.3 Gpixels/sec |
| Particle Physics | — | 42574.6 FPS |
| Stereo Matching | — | 3040 Gpixels/sec |
General
| Specification | Apple M4 Ultra GPU (80-core) | Nvidia RTX Pro 6000 Blackwell |
|---|---|---|
| Architecture | Apple M GPU | Blackwell 2.0 |
| Build | Integrated | Discrete |
| Case | Laptop | Desktop |
| GPU Codename | Custom | GB202-300-A1 |
| Nvidia Equivalent | -GeForce RTX 5080 | — |
| Purpose | Professional | Professional |
| Recommended CPU | -Apple M4 Ultraor above | -AMD Ryzen 9 9955HX3Dor above |
| Released | May 1, 2025 | March 18, 2026 |
| Segment | High-end | High-end |
| Used in CPUs | -Apple M4 Ultra | — |
| Vendor | Apple | Nvidia |
| Launch price (MSRP) | — | $8559 |
Graphics Processing Unit
| Specification | Apple M4 Ultra GPU (80-core) | Nvidia RTX Pro 6000 Blackwell |
|---|---|---|
| Base Clock | 500 MHz | 1590 MHz |
| Boost Clock | 1578 MHz | 2617 MHz |
| Compute Units (Pipelines) | 1280 | 188 |
| Instructions Per Cycle | 2 IPC | 2 IPC |
| Render Output Units (ROPs) | 320 | 190 |
| Shading Units | 10240 | 24064 |
| Texture Mapping Units (TMUs) | 640 | 752 |
| L1 Cache | — | 128KB per cluster |
| L2 Cache | — | 128MB shared |
| Ray-tracing Cores | — | 188 |
| Tensor Cores | — | 752 |
Memory
| Specification | Apple M4 Ultra GPU (80-core) | Nvidia RTX Pro 6000 Blackwell |
|---|---|---|
| Bus | 1024-bit | 512-bit |
| ECC | No | Yes |
| Memory Clock | 8533 MHz | 1750 MHz |
| Memory Type | System Shared | GDDR7 |
| Effective Memory Speed | — | 28000 Mbps |
| Memory Size | — | 96 GB |
Physical
| Specification | Apple M4 Ultra GPU (80-core) | Nvidia RTX Pro 6000 Blackwell |
|---|---|---|
| Fabrication Process | 3 nm | 5 nm |
| Interface | Custom | PCIe 5.0 x16 |
| Manufacturing | TSMC | TSMC |
| TGP | 120 W | 600 W |
| Die Size | — | 750 mm² |
| Transistor Count | — | 92.2 billion |
| Transistor Density | — | 122.93 MTr/mm² |
Raw Performance
| Specification | Apple M4 Ultra GPU (80-core) | Nvidia RTX Pro 6000 Blackwell |
|---|---|---|
| Pixel Fill Rate | 505 GPixel/s | 497 GPixel/s |
| Texture Fill Rate | 1010 GTexel/s | 1968 GTexel/s |
3D Mark
| Specification | Apple M4 Ultra GPU (80-core) | Nvidia RTX Pro 6000 Blackwell |
|---|---|---|
| Fire Strike | — | 75325 |
| Night Raid | — | 214655 |
| Port Royal | — | 39888 |
| Solar Bay | — | 255911 |
| Time Spy | — | 42614 |
| Wild Life Extreme | — | 110070 |
Games
| Specification | Apple M4 Ultra GPU (80-core) | Nvidia RTX Pro 6000 Blackwell |
|---|---|---|
| 1080p High | — | 284 |
| 1080p Ultra | — | 242 |
| 1440p Ultra | — | 196 |
| 4K Ultra | — | 118 |
| CoD: Modern Warfare III | — | 305296252176 |
| Counter-Strike 2 | — | 36226320393 |
| Cyberpunk 2077 | — | 25322116071 |
| Far Cry 6 | — | 221198185112 |
| Forza Horizon 5 | — | 309235215175 |
| Ghost of Tsushima | — | 206170158107 |
| Hogwarts Legacy | — | 23518915089 |
| Margin of Error | — | High |
| Shadow of the Tomb Raider | — | 319298224121 |
| The Witcher 3 | — | 343311216116 |
GeekBench 6 ML
| Specification | Apple M4 Ultra GPU (80-core) | Nvidia RTX Pro 6000 Blackwell |
|---|---|---|
| Backend | — | DirectML |
| Depth Estimation (Q) | — | 90853 |
| Face Detection (HP) | — | 103224 |
| Framework | — | ONNX |
| Image Classification (SP) | — | 16903 |
| Image Segmentation (HP) | — | 48518 |
| Image Super Resolution (Q) | — | 54618 |
| Machine Translation (HP) | — | 7291 |
| Object Detection (SP) | — | 26498 |
| Pose Estimation (Q) | — | 338587 |
| Style Transfer (SP) | — | 635289 |
| Text Classification (SP) | — | 3894 |
Passmark Graphics
| Specification | Apple M4 Ultra GPU (80-core) | Nvidia RTX Pro 6000 Blackwell |
|---|---|---|
| DirectX 11 | — | 339 FPS |
| DirectX 12 | — | 154 FPS |
| G2D Mark | — | 1391 |
| GPU Compute | — | 26607 Ops/s |
Looking for More Comparisons?
Browse all available head-to-head comparisons or select new devices in the comparison hub.