Apple M2 Ultra (76-GPU) Benchmark, Test and specs
Last updated:
The Apple M2 Ultra (76-GPU) has 24 CPU cores and can process 24 threads at the same time. The processor was presented in Q2/2023 and is based on the 2. Generation of the Apple M series series. In the Geekbench 5 benchmark, the Apple M2 Ultra (76-GPU) achieved a result of 1,940 points (single-core) or 27,860 points (multi-core).
| Name: | Apple M2 Ultra (76-GPU) |
|---|---|
| Family: | Apple M series (23) |
| CPU group: | Apple M2 (8) |
| Architecture: | M2 |
| Segment: | Mobile |
| Generation: | 2 |
| Predecessor: | Apple M1 Ultra (64-GPU) |
| Successor: | -- |
CPU Cores and Base Frequency
The Apple M2 Ultra (76-GPU) has 24 cores. The clock frequency of the Apple M2 Ultra (76-GPU) is 0.66 GHz (3.50 GHz). An initial performance assessment can be made using the number of CPU cores.
| CPU Cores / Threads: | 24 / 24 |
|---|---|
| Core architecture: | hybrid (big.LITTLE) |
| A-Core: | 16x Avalanche |
| B-Core: | 8x Blizzard |
| Hyperthreading / SMT: | No |
|---|---|
| Overclocking: | No |
| A-Core Frequency: | 0.66 GHz (3.50 GHz) |
| B-Core Frequency: | 0.60 GHz (2.42 GHz) |
Artificial Intelligence and Machine Learning
Processors with the support of artificial intelligence (AI) and machine learning (ML) can process many calculations, especially audio, image and video processing, much faster than classic processors. Algorithms for ML improve their performance the more data they have collected via software. ML tasks can be processed up to 10,000 times faster than with a classic processor.
| AI hardware: | Apple Neural Engine |
|---|---|
| AI specifications: | 32 Neural cores @ 31.6 TOPS |
Internal Graphics
The Apple M2 Ultra (76-GPU) has an integrated graphics that the system can use to efficiently play back videos. The Apple M2 Ultra (76-GPU) has the Apple M2 Ultra (76 Core) installed, which has 1216 streaming multiprocessors (9728 shaders).
| GPU name: | Apple M2 Ultra (76 Core) |
|---|---|
| GPU frequency: | 0.45 GHz |
| GPU (Turbo): | 1.40 GHz |
| Compute units: | 1216 |
| Shader: | 9728 |
| Hardware Raytracing: | No |
| Release date: | Q2/2023 |
| Max. displays: | 8 |
|---|---|
| Generation: | 2 |
| Direct X: | -- |
| Technology: | 5 nm |
| Max. GPU Memory: | 192 GB |
| Frame Generation: | No |
Hardware codec support
Processors with integrated graphics can process video codecs faster. Support for modern codecs can significantly increase system efficiency during video playback.
| h265 / HEVC (8 bit): | Decode / Encode |
|---|---|
| h265 / HEVC (10 bit): | Decode / Encode |
| h264: | Decode / Encode |
| VP8: | Decode |
| VP9: | Decode / Encode |
| AV1: | No |
|---|---|
| AVC: | Decode |
| VC-1: | Decode |
| JPEG: | Decode / Encode |
Memory & PCIeThe Apple M2 Ultra (76-GPU) supports a maximum of 192 GB memory. Depending on the mainboard, the processor can use a maximum of 8 (Octa Channel) memory channels. This results in a maximum bandwidth of the main memory of 819.1 GB/s. |
|
| Memory type: | Memory bandwidth: |
|---|---|
| LPDDR5-6400 | 819.1 GB/s |
| Max. Memory: | 192 GB |
| Memory channels: | 8 (Octa Channel) |
| ECC: | No |
| PCIe: | 4.0 x 64 |
| PCIe Bandwidth: | 126.0 GB/s |
Thermal ManagementThe Apple M2 Ultra (76-GPU) has a TDP of 80 W. Based on the TDP, the system manufacturer can and must adapt the cooling solution to the processor. |
|
|---|---|
| TDP (PL1 / PBP): | 80 W |
| TDP (PL2): | -- |
| TDP up: | -- |
| TDP down: | -- |
| Tjunction max.: | 100 °C |
Technical details
Modern production reduces the waste heat of a processor and increases its efficiency. The Apple M2 Ultra (76-GPU) is made in 5 nm and has 72.00 MB cache.
| Technology: | 5 nm |
|---|---|
| Chip design: | Chiplet |
| Socket: | -- |
| L2-Cache: | 72.00 MB |
| L3-Cache: | -- |
| AES-NI: | Yes |
| Operating systems: | macOS, iPadOS |
| Virtualization: | Apple Virtualization Framework |
|---|---|
| Instruction set (ISA): | Armv8.5-A (64 bit) |
| ISA extensions: | Rosetta 2 x86-Emulation |
| Release date: | Q2/2023 |
| Release price: | -- |
| Part Number: | -- |
| Documents: | -- |
24C 24T @ 0.66 GHz (3.50 GHz)
buy now on Amazon and save!
Rate this processor
Benchmark results
The benchmark results for the Apple M2 Ultra (76-GPU) have been carefully checked by us. We only publish benchmark results that have been created by us or that have been submitted by a visitor and then checked by a team member. All results are based on and fullfill our benchmark guidelines.
Cinebench 2024 (Single-Core)
The Cinebench 2024 benchmark is based on the Redshift rendering engine, which is also used in Maxon's 3D program Cinema 4D. The benchmark runs are each 10 minutes long to test whether the processor is limited by its heat generation.
|
Intel Core i9-13900K
24C 32T @ 5.80 GHz |
||
|
Intel Core i7-14700K
20C 28T @ 5.60 GHz |
||
|
|
Intel Core i7-14700KF
20C 28T @ 5.60 GHz |
||
|
Apple M2 Ultra (76-GPU)
24C 24T @ 3.50 GHz |
||
|
|
Apple M2 Ultra (60-GPU)
24C 24T @ 3.50 GHz |
||
|
|
Intel Core i7-13700KF
16C 24T @ 5.40 GHz |
||
|
|
Intel Core i7-13700K
16C 24T @ 5.40 GHz |
||
Cinebench 2024 (Multi-Core)
The Multi-Core test of the Cinebench 2024 benchmark uses all cpu cores to render using the Redshift rendering engine, which is also used in Maxons Cinema 4D. The benchmark run is 10 minutes long to test whether the processor is limited by its heat generation.
|
|
Intel Core i7-14700K
20C 28T @ 5.60 GHz |
||
|
|
Intel Core i7-14700KF
20C 28T @ 5.60 GHz |
||
|
AMD Ryzen Threadripper 3960X
24C 48T @ 4.50 GHz |
||
|
|
Apple M2 Ultra (76-GPU)
24C 24T @ 3.50 GHz |
||
|
|
Apple M2 Ultra (60-GPU)
24C 24T @ 3.50 GHz |
||
|
Intel Xeon W-3175X
28C 56T @ 3.80 GHz |
||
|
|
Intel Core i9-13900HX
24C 32T @ 5.40 GHz |
||
Cinebench R23 (Single-Core)
Cinebench R23 is the successor of Cinebench R20 and is also based on the Cinema 4 Suite. Cinema 4 is a worldwide used software to create 3D forms. The single-core test only uses one CPU core, the amount of cores or hyperthreading ability doesn't count.
|
Intel Core i5-1345U
10C 12T @ 4.70 GHz |
||
|
|
Intel Core i5-1340P
12C 16T @ 4.60 GHz |
||
|
|
Intel Core i5-1335U
10C 12T @ 4.60 GHz |
||
|
|
Apple M2 Ultra (76-GPU)
24C 24T @ 3.50 GHz |
||
|
|
Apple M2 Ultra (60-GPU)
24C 24T @ 3.50 GHz |
||
|
|
Apple M2 Max (30-GPU)
12C 12T @ 3.50 GHz |
||
|
|
Apple M2 Pro (10-CPU 16-GPU)
10C 10T @ 3.50 GHz |
||
Cinebench R23 (Multi-Core)
Cinebench R23 is the successor of Cinebench R20 and is also based on the Cinema 4 Suite. Cinema 4 is a worldwide used software to create 3D forms. The multi-core test involves all CPU cores and taks a big advantage of hyperthreading.
|
AMD Ryzen 9 7900
12C 24T @ 4.70 GHz |
||
|
AMD Ryzen 9 PRO 7945
12C 24T @ 4.70 GHz |
||
|
|
AMD Ryzen 9 5950X
16C 32T @ 4.20 GHz |
||
|
|
Apple M2 Ultra (76-GPU)
24C 24T @ 3.50 GHz |
||
|
|
Apple M2 Ultra (60-GPU)
24C 24T @ 3.50 GHz |
||
|
|
Intel Core i9-7980XE
18C 36T @ 3.50 GHz |
||
|
|
Intel Xeon W-3275
28C 56T @ 3.40 GHz |
||
Geekbench 5, 64bit (Single-Core)
Geekbench 5 is a cross plattform benchmark that heavily uses the systems memory. A fast memory will push the result a lot. The single-core test only uses one CPU core, the amount of cores or hyperthreading ability doesn't count.
|
AMD Ryzen 5 8600G
6C 12T @ 5.00 GHz |
||
|
AMD Ryzen 5 PRO 8600G
6C 12T @ 5.00 GHz |
||
|
|
Intel Core i7-13850HX
20C 28T @ 5.30 GHz |
||
|
|
Apple M2 Ultra (76-GPU)
24C 24T @ 3.50 GHz |
||
|
|
Apple M2 Ultra (60-GPU)
24C 24T @ 3.50 GHz |
||
|
|
Intel Core i7-13800H
14C 20T @ 5.20 GHz |
||
|
|
Intel Core i5-14600
14C 20T @ 5.20 GHz |
||
Geekbench 5, 64bit (Multi-Core)
Geekbench 5 is a cross plattform benchmark that heavily uses the systems memory. A fast memory will push the result a lot. The multi-core test involves all CPU cores and taks a big advantage of hyperthreading.
|
AMD EPYC 7443P
24C 48T @ 3.60 GHz |
||
|
|
Intel Core i9-14900KS
24C 32T @ 5.90 GHz |
||
|
|
AMD EPYC 74F3
24C 48T @ 3.80 GHz |
||
|
|
Apple M2 Ultra (76-GPU)
24C 24T @ 3.50 GHz |
||
|
|
Apple M2 Ultra (60-GPU)
24C 24T @ 3.50 GHz |
||
|
|
Intel Xeon W-3375
38C 72T @ 3.20 GHz |
||
|
|
Intel Core i9-14900KF
24C 32T @ 5.70 GHz |
||
Geekbench 6 (Single-Core)
Geekbench 6 is a benchmark for modern computers, notebooks and smartphones. What is new is an optimized utilization of newer CPU architectures, e.g. based on the big.LITTLE concept and combining CPU cores of different sizes. The single-core benchmark only evaluates the performance of the fastest CPU core, the number of CPU cores in a processor is irrelevant here.
|
|
Intel Core i7-14650HX
16C 24T @ 5.20 GHz |
||
|
|
Intel Core i7-13850HX
20C 28T @ 5.30 GHz |
||
|
|
AMD Ryzen 9 7940HX
16C 32T @ 5.20 GHzNot verified |
||
|
|
Apple M2 Ultra (76-GPU)
24C 24T @ 3.50 GHz |
||
|
|
Apple M2 Ultra (60-GPU)
24C 24T @ 3.50 GHz |
||
|
|
Apple M2 Max (30-GPU)
12C 12T @ 3.50 GHz |
||
|
|
Apple M2 Pro (10-CPU 16-GPU)
10C 10T @ 3.50 GHz |
||
Geekbench 6 (Multi-Core)
Geekbench 6 is a benchmark for modern computers, notebooks and smartphones. What is new is an optimized utilization of newer CPU architectures, e.g. based on the big.LITTLE concept and combining CPU cores of different sizes. The multi-core benchmark evaluates the performance of all of the processor's CPU cores. Virtual thread improvements such as AMD SMT or Intel's Hyper-Threading have a positive impact on the benchmark result.
|
|
AMD EPYC 9654
96C 192T @ 3.55 GHz |
||
|
|
Intel Core i9-14900K
24C 32T @ 5.70 GHz |
||
|
|
Intel Core i9-14900KF
24C 32T @ 5.70 GHz |
||
|
|
Apple M2 Ultra (76-GPU)
24C 24T @ 3.50 GHz |
||
|
|
Apple M2 Ultra (60-GPU)
24C 24T @ 3.50 GHz |
||
|
|
Intel Xeon w9-3475X
36C 72T @ 3.00 GHz |
||
|
|
Apple M3 Max (16-CPU 40-GPU)
16C 16T @ 3.60 GHz |
||
iGPU - FP32 Performance (Single-precision GFLOPS)
The theoretical computing performance of the internal graphics unit of the processor with simple accuracy (32 bit) in GFLOPS. GFLOPS indicates how many billion floating point operations the iGPU can perform per second.
|
|
Apple M2 Ultra (76-GPU)
Apple M2 Ultra (76 Core) @ 1.40 GHz |
||
|
|
Apple M2 Ultra (60-GPU)
Apple M2 Ultra (60 Core) @ 1.40 GHz |
||
|
|
Apple M1 Ultra (64-GPU)
Apple M1 Ultra (64 Core) @ 1.30 GHz |
||
|
|
Apple M1 Ultra (48-GPU)
Apple M1 Ultra (48 Core) @ 1.30 GHz |
||
|
|
Apple M3 Max (16-CPU 40-GPU)
Apple M3 Max (40 Core) @ 1.40 GHz |
||
|
|
Apple M2 Max (38-GPU)
Apple M2 Max (38 Core) @ 1.40 GHz |
||
|
Microsoft XBox Series X
AMD Custom Radeon Graphics (XBox Series X) @ 1.83 GHz |
||
Estimated results for PassMark CPU Mark
Some of the CPUs listed below have been benchmarked by CPU-monkey. However the majority of CPUs have not been tested and the results have been estimated by a CPU-monkey’s secret proprietary formula. As such they do not accurately reflect the actual Passmark CPU mark values and are not endorsed by PassMark Software Pty Ltd.
|
|
Intel Core i7-14700
20C 28T @ 5.30 GHz |
||
|
|
Intel Core i7-14700F
20C 28T @ 5.30 GHz |
||
|
|
Intel Core i9-13980HX
24C 32T @ 4.90 GHz |
||
|
|
Apple M2 Ultra (76-GPU)
24C 24T @ 3.50 GHz |
||
|
|
Apple M2 Ultra (60-GPU)
24C 24T @ 3.50 GHz |
||
|
|
Intel Xeon W-3345
24C 48T @ 3.70 GHz |
||
|
|
AMD EPYC 7402
24C 48T @ 3.20 GHz |
||
CPU performance per watt (efficiency)
Efficiency of the processor under full load in the Cinebench R23 (multi-core) benchmark. The benchmark result is divided by the average energy required (CPU package power in watts). The higher the value, the more efficient the CPU is under full load.
|
|
Apple M2 Max (30-GPU)
14,855 CB R23 MC @ 40 W |
||
|
|
Apple M2 Max (38-GPU)
14,855 CB R23 MC @ 40 W |
||
|
|
Apple M2 Pro (12-CPU 19-GPU)
14,855 CB R23 MC @ 40 W |
||
|
|
Apple M2 Ultra (76-GPU)
28,570 CB R23 MC @ 80 W |
||
|
|
Apple M2 Ultra (60-GPU)
28,570 CB R23 MC @ 80 W |
||
|
|
AMD Ryzen 9 7945HX3D
33,450 CB R23 MC @ 104 W |
||
|
|
AMD Ryzen 9 7900
28,905 CB R23 MC @ 90 W |
||
Performance for Artificial Intelligence (AI) and Machine Learning (ML)
Processors with the support of artificial intelligence (AI) and machine learning (ML) can process many calculations, especially audio, image and video processing, much faster than classic processors. The performance is given in the number (trillions) of arithmetic operations per second (TOPS).
|
MediaTek Dimensity 9300
8C 8T @ 3.25 GHz |
||
|
Qualcomm Snapdragon 888+
8C 8T @ 3.00 GHz |
||
|
|
Apple M2 Ultra (60-GPU)
24C 24T @ 0.66 GHz |
||
|
|
Apple M2 Ultra (76-GPU)
24C 24T @ 0.66 GHz |
||
|
|
Qualcomm Snapdragon 8 Gen 2 for Galaxy
8C 8T @ 3.36 GHz |
||
|
|
Qualcomm Snapdragon 8 Gen 2
8C 8T @ 3.36 GHz |
||
|
|
Qualcomm Snapdragon 8 Gen 1
8C 8T @ 3.00 GHz |
||
Benchmarks
Cinebench 2024 (SC)
272 entries
272 entries
Cinebench 2024 (MC)
271 entries
271 entries
Cinebench R23 (SC)
586 entries
586 entries
Cinebench R23 (MC)
565 entries
565 entries
Geekbench 5 (SC)
2,488 entries
2,488 entries
Geekbench 5 (MC)
2,461 entries
2,461 entries
Geekbench 6 (SC)
1,755 entries
1,755 entries
Geekbench 6 (MC)
1,703 entries
1,703 entries
FP32 SP (iGPU)
2,026 entries
2,026 entries
PassMark CPU-Mark
2,392 entries
2,392 entries
CPU performance per watt (efficiency)
109 entries
109 entries
AI / ML Performance
116 entries
116 entries
Description of the processor
The Apple M2 Ultra (variant with 76 GPU cores) is a 24-core processor and thus the maximum configuration of the Apple M2 series. The processor is available in two variants: one with 60 GPU cores and one with 76 GPU cores. Except for the graphics unit, the two Apple M2 processors are identical.With the Apple M2 Ultra, Apple also relies on a hybrid core structure of 16 large and fast "Avalanche" CPU cores, which are supplemented by 8 smaller and more efficient "Blizzard" CPU cores. The clock frequency of the processor is 3.5 GHz, the smaller efficiency cores clock at a maximum of 2.8 GHz.
Like all current Apple processors, the Apple M2 Ultra also features the Apple Neural Engine. In the Apple M2 Ultra, the AI engine with 32 cores (31.6 TOPS) is twice as powerful as in the Apple M2 Max. The Apple Neural Engine accelerates AI tasks in hardware. This can, for example, significantly speed up the computing time for image and video processing.
Apples currently fastest graphics unit is installed in the Apple M2 Ultra with 76 GPU cores. With 9728 texture shaders and 1216 SM processors, the 76-core expansion stage has a theoretical computing power of almost 27 TFLOPS and is thus roughly at the level of an NVIDIA GeForce RTX 3080 (30 TFLOPS). For comparison: NVIDIAs current top model, the NVIDIA GeForce RTX 4090, has an FP32 speed of 83 TFLOPS.
The Apple M2 Ultra can be equipped with a maximum of 192 GB LPDDR5 RAM (LPDDR5-6400) and achieves a maximum bandwidth of 820 GB per second via 8 memory channels. This is an absolute record value for a processor, but it is also required in order not to slow down the fast integrated graphics unit of the Apple M2 Ultra.
Apple does not officially state the TDP of the processor, but it should be around 90 watts. The Apple M2 Ultra is still manufactured in a 5 nm process. Only the Apple M3 is to be manufactured in an even finer 3 nm manufacturing process.
Popular comparisons
back to index
