Apple M4 Ultra GPU (80-core) vs M1 Pro GPU (16-core)
We compared two integrated laptop professional GPUs: the Apple M4 Ultra GPU (80-core) with 1280 pipelines and 10240 shaders against the 3 years and 7 months older M1 Pro GPU (16-core) that utilizes 256 pipelines and 2048 shaders. Here you will find complete details about specs, efficiency, performance tests, and more.
Please note that the tests on the Apple M4 Ultra GPU (80-core) are done on an engineering sample provided by our insiders. The data will be more accurate after the final version of this GPU is available.
Key differences
Reasons to consider the Apple M4 Ultra GPU (80-core)
- Performs significantly better (up to 6.5x) in 3DMark Steel Nomad Lite
- 6.1x higher maximum theoretical performance (32.3 vs 5.3 TFLOPS)
- Manufactured using a more efficient 3 nm process technology
- Achieves 5.2x more points in the GeekBench 6 Compute test (218K vs 42K)
- Has 5x more shading units (10240 vs 2048)
Benchmarks
Graphics cardsโ performance in recent benchmarking apps3D Mark
Steel Nomad Lite Score
| Solar Bay | - | 12485 |
| Wild Life Extreme | - | 9873 |
GeekBench 6 OpenCL
GB6 Compute Score
| Background Blur | - | 74.5 img/sec |
| Face Detection | - | 47.3 img/sec |
| Horizon Detection | - | 1.73 Gpixels/sec |
| Edge Detection | - | 3.32 Gpixels/sec |
| Gaussian Blur | - | 1.75 Gpixels/sec |
| Feature Matching | - | 0.47 Gpixels/sec |
| Stereo Matching | - | 131.2 Gpixels/sec |
| Particle Physics | - | 5143 FPS |
| API | OpenCL | OpenCL |
Cinebench 2024 GPU
Cinebench 2024 GPU
Recent User Tests
Apple M4 Ultra GPU (80-core)
No benchmark results yet
Apple M1 Pro GPU (16-core)
| Date | Benchmark | Result |
|---|---|---|
| ๐ 2026-03-27 (bbffx) | Geekbench 6 OpenCL | 43467 |
| ๐ 2026-03-27 (bbffx) | Cinebench 2024 | 2423 |
| ๐ 2026-03-27 (bbffx) | Steel Nomad Light | 4048 |
| ๐ 2025-02-21 (serhii) | Geekbench 6 OpenCL | 40932 |
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) | - | 4112 |
| Image Segmentation (HP) | - | 7488 |
| Image Super Resolution (Q) | - | 9882 |
| Face Detection (HP) | - | 14702 |
| Pose Estimation (Q) | - | 33560 |
| Text Classification (SP) | - | 2289 |
| Machine Translation (HP) | - | 1629 |
| Object Detection (SP) | - | 3633 |
| Depth Estimation (Q) | - | 16986 |
| Style Transfer (SP) | - | 53586 |
| Framework | - | Core ML |
| Backend | - | GPU |
Specifications
Technical specifications of Apple M4 Ultra GPU (80-core) and M1 Pro GPU (16-core)General
| Vendor | Apple | Apple |
| Build | Integrated | Integrated |
| Released | May 1, 2025 | October 18, 2021 |
| Case | Laptop | Laptop |
| Purpose | Professional | Professional |
| Segment | High-end | Mid-range |
| Architecture | Apple M GPU | Apple M GPU |
| GPU Codename | Custom | - |
| Rival Equivalent | - GeForce RTX 5080 | - GeForce RTX 3050 Laptop |
| Successor | - | - Apple M5 Pro GPU (20-core) |
| Recommended CPU | - Apple M4 Ultra or above | - Apple M1 Pro or above |
| Used in CPUs | - Apple M4 Ultra | - Apple M1 Pro |
Laptop GPU ranking (#79th place)
Graphics Processing Unit
| Base Clock | 500 MHz | 450 MHz |
| Boost Clock | 1578 MHz | 1296 MHz |
| Shading Units | 10240 | 2048 |
| Texture Mapping Units (TMUs) | 640 | 128 |
| Render Output Units (ROPs) | 320 | 64 |
| Compute Units (Pipelines) | 1280 | 256 |
| Ray-tracing Cores | - | No |
| Instructions Per Cycle | 2 IPC | 2 IPC |
Raw Performance
| Pixel Fill Rate | 505 GPixel/s | 83 GPixel/s |
| Texture Fill Rate | 1010 GTexel/s | 166 GTexel/s |
FLOPS (FP32)
Physical
| Interface | Custom | Custom |
| TGP | 120 W | 30 W |
| Manufacturing | TSMC | TSMC |
| Fabrication Process | 3 nm | 5 nm |
| Transistor Count | - | 23.2 billion |
| Max. Temperature | - | 94ยฐC |
Memory
| Memory Type | System Shared | System Shared |
| Memory Clock | 8533 MHz | 6400 MHz |
| Effective Memory Speed | - | 12800 Mbps |
| Bus | 1024-bit | 256-bit |
| ECC | No | No |
Memory Bandwidth
API
| Ray Tracing | Yes | No |
| DLSS | No | No |
Cast your vote
Total votes: < 1