Apple M3 Pro GPU (18-core) vs M1 Max GPU (32-core)
We compared two integrated laptop professional GPUs: the Apple M3 Pro GPU (18-core) with 288 pipelines and 2304 shaders against the 2 years and 1 month older M1 Max GPU (32-core) that utilizes 512 pipelines and 4096 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
Key differences
Reasons to consider the Apple M3 Pro GPU (18-core)
- Manufactured using a more efficient 3 nm process technology
Reasons to consider the Apple M1 Max GPU (32-core)
- Performs better (up to 21%) in 3DMark Steel Nomad Lite
- 66% higher maximum theoretical performance (10.6 vs 6.4 TFLOPS)
- Has 2.7x higher memory bandwidth: 409.6 vs 153.6 GB/s
- Achieves 67% more points in the GeekBench 6 Compute test (71K vs 43K)
- Has 78% more shading units (4096 vs 2304)
Benchmarks
Graphics cards’ performance in recent benchmarking apps3D Mark
Steel Nomad Lite Score
| Solar Bay | 22637 | 22586 |
| Wild Life Extreme | 14021 | 19880 |
GeekBench 6 OpenCL
GB6 Compute Score
| Background Blur | 82.1 img/sec | 120.7 img/sec |
| Face Detection | 53.6 img/sec | 78.9 img/sec |
| Horizon Detection | 1.82 Gpixels/sec | 3.18 Gpixels/sec |
| Edge Detection | 2.59 Gpixels/sec | 6.15 Gpixels/sec |
| Gaussian Blur | 1.84 Gpixels/sec | 3.44 Gpixels/sec |
| Feature Matching | 0.48 Gpixels/sec | 0.62 Gpixels/sec |
| Stereo Matching | 148.6 Gpixels/sec | 229.8 Gpixels/sec |
| Particle Physics | 4893.9 FPS | 8822.9 FPS |
| API | OpenCL | OpenCL |
Cinebench 2024 GPU
Cinebench 2024 GPU
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) | 5578 | 5466 |
| Image Segmentation (HP) | 9924 | 12415 |
| Image Super Resolution (Q) | 13323 | 15930 |
| Face Detection (HP) | 19900 | 23179 |
| Pose Estimation (Q) | 35730 | 58520 |
| Text Classification (SP) | 2904 | 2299 |
| Machine Translation (HP) | 4985 | 1713 |
| Object Detection (SP) | 5496 | 5388 |
| Depth Estimation (Q) | 23344 | 26458 |
| Style Transfer (SP) | 78342 | 98278 |
| Framework | Core ML | Core ML |
| Backend | GPU | GPU |
Specifications
Technical specifications of Apple M3 Pro GPU (18-core) and M1 Max GPU (32-core)General
| Vendor | Apple | Apple |
| Build | Integrated | Integrated |
| Released | October 31, 2023 | October 18, 2021 |
| Case | Laptop | Laptop |
| Purpose | Professional | Professional |
| Segment | Mid-range | High-end |
| Architecture | Apple M GPU | Apple M GPU |
| GPU Codename | Custom | - |
| Rival Equivalent | - GeForce RTX 4060 Laptop | - GeForce RTX 3060 Laptop |
| Successor | - Apple M5 Pro GPU (20-core) | - Apple M5 Max GPU (40-core) |
| Recommended CPU | - Apple M3 Pro or above | - Apple M1 Max or above |
| Used in CPUs | - Apple M3 Pro | - Apple M1 Max |
Laptop GPU ranking (60th and 53rd place)
Graphics Processing Unit
| Base Clock | 500 MHz | 450 MHz |
| Boost Clock | 1380 MHz | 1296 MHz |
| Shading Units | 2304 | 4096 |
| Texture Mapping Units (TMUs) | 144 | 256 |
| Render Output Units (ROPs) | 72 | 128 |
| Compute Units (Pipelines) | 288 | 512 |
| Ray-tracing Cores | - | No |
| Instructions Per Cycle | 2 IPC | 2 IPC |
Raw Performance
| Pixel Fill Rate | 99 GPixel/s | 166 GPixel/s |
| Texture Fill Rate | 199 GTexel/s | 332 GTexel/s |
FLOPS (FP32)
Physical
| Interface | Custom | Custom |
| TGP | 30 W | - |
| Manufacturing | TSMC | TSMC |
| Fabrication Process | 3 nm | 5 nm |
| Transistor Count | 25.2 billion | 46.4 billion |
| Max. Temperature | 100°C | 94°C |
Memory
| Memory Type | System Shared | System Shared |
| Memory Clock | 6400 MHz | 6400 MHz |
| Effective Memory Speed | 12800 Mbps | 12800 Mbps |
| Bus | 192-bit | 512-bit |
| ECC | No | No |
Memory Bandwidth
API
| Ray Tracing | Yes | No |
| DLSS | No | No |
Cast your vote
Total votes: 12