Apple M4 Pro GPU (20-core) vs M3 GPU (10-core)
We compared two integrated laptop professional GPUs: the Apple M4 Pro GPU (20-core) with 320 pipelines and 2560 shaders against the 1 year older M3 GPU (10-core) that utilizes 160 pipelines and 1280 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 M4 Pro GPU (20-core)
- Performs significantly better (up to 2.3x) in 3DMark Steel Nomad Lite
- 2.3x higher maximum theoretical performance (8.1 vs 3.5 TFLOPS)
- Has 2.7x higher memory bandwidth: 273 vs 102.4 GB/s
- Achieves 2.3x more points in the GeekBench 6 Compute test (70K vs 30K)
- Has 2x more shading units (2560 vs 1280)
Benchmarks
Graphics cards’ performance in recent benchmarking apps3D Mark
Steel Nomad Lite Score
| Solar Bay | 33418 | 13531 |
| Wild Life Extreme | 19535 | 8237 |
GeekBench 6 OpenCL
GB6 Compute Score
| Background Blur | 125.1 img/sec | 60.3 img/sec |
| Face Detection | 80 img/sec | 40.8 img/sec |
| Horizon Detection | 2.88 Gpixels/sec | 1.21 Gpixels/sec |
| Edge Detection | 4.22 Gpixels/sec | 1.64 Gpixels/sec |
| Gaussian Blur | 2.98 Gpixels/sec | 1.2 Gpixels/sec |
| Feature Matching | 0.8 Gpixels/sec | 0.37 Gpixels/sec |
| Stereo Matching | 235 Gpixels/sec | 102.6 Gpixels/sec |
| Particle Physics | 9516.1 FPS | 3293.3 FPS |
| API | OpenCL | OpenCL |
Cinebench 2024 GPU
Cinebench 2024 GPU
Blender
Blender GPU
Recent User Tests
Apple M4 Pro GPU (20-core)
No benchmark results yet
Apple M3 GPU (10-core)
| Date | Benchmark | Result |
|---|---|---|
| 📘 2025-04-24 (Alex) | Geekbench 6 OpenCL | 31004 |
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) | 8046 | 1579 |
| Image Segmentation (HP) | 15443 | 2944 |
| Image Super Resolution (Q) | 20807 | 4425 |
| Face Detection (HP) | 30860 | 6661 |
| Pose Estimation (Q) | 56783 | 15020 |
| Text Classification (SP) | 2888 | 2984 |
| Machine Translation (HP) | 6167 | 5876 |
| Object Detection (SP) | 7592 | 1826 |
| Depth Estimation (Q) | 33607 | 9339 |
| Style Transfer (SP) | 126726 | 22291 |
| Framework | Core ML | Core ML |
| Backend | GPU | CPU |
Specifications
Technical specifications of Apple M4 Pro GPU (20-core) and M3 GPU (10-core)General
| Vendor | Apple | Apple |
| Build | Integrated | Integrated |
| Released | October 29, 2024 | October 31, 2023 |
| Case | Laptop | Laptop |
| Purpose | Professional | Professional |
| Segment | Mid-range | Mid-range |
| Architecture | Apple M GPU | Apple M GPU |
| GPU Codename | Custom | - |
| Rival Equivalent | - GeForce RTX 4060 Laptop | - GeForce RTX 4060 Laptop |
| Successor | - Apple M5 Pro GPU (20-core) | - Apple M5 GPU (10-Core) |
| Recommended CPU | - Apple M4 Pro (14-Core) or above | - Apple M3 or above |
| Used in CPUs | - Apple M4 Pro (14-Core) | - Apple M3 |
Laptop GPU ranking (37th and 80th place)
Graphics Processing Unit
| Base Clock | 500 MHz | 500 MHz |
| Boost Clock | 1578 MHz | 1380 MHz |
| Shading Units | 2560 | 1280 |
| Texture Mapping Units (TMUs) | 160 | 80 |
| Render Output Units (ROPs) | 80 | 40 |
| Compute Units (Pipelines) | 320 | 160 |
| Instructions Per Cycle | 2 IPC | 2 IPC |
Raw Performance
| Pixel Fill Rate | 126 GPixel/s | 55 GPixel/s |
| Texture Fill Rate | 252 GTexel/s | 110 GTexel/s |
FLOPS (FP32)
Physical
| Interface | Custom | Custom |
| TGP | 26 W | 15 W |
| Manufacturing | TSMC | TSMC |
| Fabrication Process | 3 nm | 3 nm |
| Transistor Count | - | 13.8 billion |
| Max. Temperature | 100°C | 100°C |
Memory
| Memory Type | System Shared | System Shared |
| Memory Clock | 8533 MHz | 6400 MHz |
| Effective Memory Speed | 15000 Mbps | 12800 Mbps |
| Bus | 192-bit | 128-bit |
| ECC | No | No |
Memory Bandwidth
API
| Ray Tracing | Yes | Yes |
| DLSS | No | No |
Cast your vote
Total votes: 1