Verdict: GMKtec EVO-X2 impresses as an all-rounder for gaming & professional tasks
The GMKtec EVO-X2 presents itself in the test as a compact PC with impressive performance. The combination of AMD Strix Halo APU with Radeon RX 8060S, 64 GB RAM and fast 1 TB SSD ensures strong performance in games, 3D rendering and workstation tasks. With 64 GB RAM, the GMKtec EVO-X2 offers slightly fewer options when it comes to local AI capabilities. GMKtec also offers alternative configurations with 96 GB RAM or 128 GB RAM. However, there is no change in the pure performance of the CPU and GPU, as the direct comparison with the Bosgame M5 in this test shows.
Despite its high computing power, the system remains comparatively cool and quiet under full load. The workmanship is excellent and the variety of connections offers all the options for most applications. Thanks to the option of assigning up to 48 GB of RAM to the graphics unit, the EVO-X2 offers an advantage over many competitors with dedicated graphics adapters. The only criticism is the complicated design, which makes maintenance work unnecessarily difficult, and the limited fan control options. Overall, the GMKtec EVO-X2 delivers a great result, which is on a par with the Bosgame M5 and is also one of the most exciting mini PCs of 2025.
Pros
Cons
Price and availability
The GMKtec EVO-X2 can be ordered directly from the manufacturer for $1,499 USD. The model with 128 GB RAM costs $1,999 USD, although the EVO-X2 with 96 GB RAM costs $1,799 USD. Amazon also offers the EVO-X2 in our test configuration for $1,499 USD.
The GMKtec EVO-X2 is currently one of the most interesting mini PCs in its class. It is powered by an AMD Ryzen AI Max+ 395, which in combination with the AMD Radeon 8060S has an enormously powerful iGPU. Combined with up to 64 GB DDR5 RAM and a fast NVMe SSD, this configuration of the EVO-X2 provides a great basis for office, multimedia and a variety of creative tasks.
Specifications
Case & features
The compact housing of the GMKtec EVO-X2 has a modern look thanks to its clean finish and discreet ventilation openings. Although it is slightly larger than most mini PCs due to the powerful hardware, it does not appear bulky. The material mix of plastic and high-quality, tactile aluminum in silver gives the device a high-quality feel.
The EVO-X2 is also flexible when it comes to connectivity: several USB-A and USB-C ports with USB4 speed, HDMI and DisplayPort enable the control of up to four monitors. There is also a 2.5 G LAN connection and Wi-Fi 7 in combination with Bluetooth 5.4, which means the mini PC is technically up to date.
ADATA SE920 USB4-Speed
Connectivity - USB 3.x / USB4 / TB - AS SSD Seq Read Test (1GB) | |
Minisforum AI X1 Pro | |
BOSGAME M5 | |
GMK EVO-T1 | |
GMK EVO-X2 | |
HP Z2 Mini G1a | |
Average of class Mini PC (423 - 3906, n=29, last 2 years) | |
Geekom Mini IT15, Ultra 9 285H |
SD card reader
SD Card Reader | |
average JPG Copy Test (av. of 3 runs) | |
BOSGAME M5 (AV Pro SD microSD 128 GB V60) | |
GMK EVO-X2 (AV Pro SD microSD 128 GB V60) | |
Geekom Mini IT15, Ultra 9 285H (AV Pro SD microSD 128 GB V60) | |
Average of class Mini PC (15.1 - 214, n=26, last 2 years) | |
Minisforum AI X1 Pro (AV Pro SD microSD 128 GB V60) | |
maximum AS SSD Seq Read Test (1GB) | |
BOSGAME M5 (AV Pro SD microSD 128 GB V60) | |
GMK EVO-X2 (AV Pro SD microSD 128 GB V60) | |
Geekom Mini IT15, Ultra 9 285H (AV Pro SD microSD 128 GB V60) | |
Average of class Mini PC (18.2 - 241, n=26, last 2 years) | |
Minisforum AI X1 Pro (AV Pro SD microSD 128 GB V60) |
Accessories
The scope of delivery includes the device and the 230-watt power supply unit. The manufacturer also includes an HDMI cable and a short quick start guide. The GMK AI Assistant is pre-installed on the mini PC, which enables the execution of local AI instances.
Maintenance
The GMKtec EVO-X2 can certainly be opened. However, the screws are located under the glued-on rubber feet, which is not exactly an advantage. After removing the screws, the two metal covers can be easily removed. The EVO-X2 does not offer many upgrade options anyway. Only an additional mass storage device (M.2-2280) can be installed.
Sustainability
GMKtec does not provide any information on sustainability measures in the production process. The packaging consists mainly of cardboard and contains foam elements to protect the device.
Performance
Test conditions
The tests relevant to the evaluation were carried out in performance mode. The GMKtec EVO-X2 also offers a Balanced and Quiet mode, which restrict the processor's power limit. The performance profile can be selected either in the BIOS or via a button on the device. The iGPU is already allocated 32 GB of RAM ex works. This can be increased to a maximum of 48 GB, leaving 16 GB available for the rest of the system.
Processor
The GMKtec EVO-X2 is, just like the Bosgame M5 or the HP Z2 Mini G1a is equipped with the AMD Ryzen AI Max+ 395 - currently one of the most powerful mobile processors from AMD. In performance mode, the SoC works with up to 140 watts for short periods, while the power consumption levels off at 120 watts under prolonged load. Despite a higher power limit of up to 160 watts in the HP Z2 Mini G1a, the EVO-X2 delivers a slightly better overall performance in the overall rating. Other systems cannot keep up with the performance combination of GMKtec EVO-X2 and Ryzen AI Max+ 395; even the Minisforum AI X1 Pro lags behind by around 10 percentage points.
Power Mode | Cinebench R15 1T | Cinebench R15 nT | Cinebench R20 1T | Cinebench R20 nT | Cinebench R23 1T | Cinebench R23 nT | Cinebench 2024 1T | Cinebench 2024 nT |
---|---|---|---|---|---|---|---|---|
Performance (140W/120W) | 314 | 5,575 | 798 | 13,882 | 2,043 | 36,951 | 115 | 1,885 |
Balanced (120W/85W) | 312 | 4,919 | 798 | 12,686 | 2,039 | 31,870 | 114 | 1,720 |
Quiet (100W/54W) | 314 | 3,829 | 794 | 10,087 | 2,042 | 25,773 | 115 | 1,443 |
Performance rating - Percent | |
Minisforum G7 PT | |
BOSGAME M5 | |
GMK EVO-X2 | |
HP Z2 Mini G1a -2! | |
Minisforum AtomMan G7 Ti | |
Minisforum AI X1 Pro | |
GMK EVO-T1 | |
Geekom Mini IT15, Ultra 9 285H |
Cinebench R10 | |
Rendering Single CPUs 64Bit | |
Geekom Mini IT15, Ultra 9 285H | |
GMK EVO-T1 | |
Minisforum AtomMan G7 Ti | |
BOSGAME M5 | |
GMK EVO-X2 | |
HP Z2 Mini G1a | |
Minisforum AI X1 Pro | |
Minisforum G7 PT | |
Rendering Multiple CPUs 64Bit | |
Minisforum G7 PT | |
GMK EVO-T1 | |
BOSGAME M5 | |
Geekom Mini IT15, Ultra 9 285H | |
GMK EVO-X2 | |
HP Z2 Mini G1a | |
Minisforum AtomMan G7 Ti | |
Minisforum AI X1 Pro |
WinRAR - Result | |
Minisforum G7 PT | |
Minisforum AtomMan G7 Ti | |
GMK EVO-X2 | |
BOSGAME M5 | |
Geekom Mini IT15, Ultra 9 285H | |
GMK EVO-T1 | |
Minisforum AI X1 Pro | |
HP Z2 Mini G1a |
3DMark 11 - 1280x720 Performance Physics | |
Minisforum G7 PT | |
GMK EVO-T1 | |
Minisforum AtomMan G7 Ti | |
BOSGAME M5 | |
GMK EVO-X2 | |
Geekom Mini IT15, Ultra 9 285H | |
Minisforum AI X1 Pro | |
HP Z2 Mini G1a |
HWBOT x265 Benchmark v2.2 - 4k Preset | |
Minisforum G7 PT | |
BOSGAME M5 | |
GMK EVO-X2 | |
HP Z2 Mini G1a | |
Minisforum AI X1 Pro | |
Minisforum AtomMan G7 Ti | |
GMK EVO-T1 | |
Geekom Mini IT15, Ultra 9 285H |
R Benchmark 2.5 - Overall mean | |
GMK EVO-T1 | |
Geekom Mini IT15, Ultra 9 285H | |
BOSGAME M5 | |
Minisforum G7 PT | |
HP Z2 Mini G1a | |
GMK EVO-X2 | |
Minisforum AtomMan G7 Ti | |
Minisforum AI X1 Pro |
LibreOffice - 20 Documents To PDF | |
HP Z2 Mini G1a | |
GMK EVO-T1 | |
Minisforum AI X1 Pro | |
BOSGAME M5 | |
Minisforum G7 PT | |
GMK EVO-X2 | |
Minisforum AtomMan G7 Ti | |
Geekom Mini IT15, Ultra 9 285H |
Super Pi mod 1.5 XS 1M - 1M | |
GMK EVO-X2 | |
BOSGAME M5 | |
HP Z2 Mini G1a | |
Minisforum AI X1 Pro | |
GMK EVO-T1 | |
Geekom Mini IT15, Ultra 9 285H | |
Minisforum G7 PT | |
Minisforum AtomMan G7 Ti |
Super Pi mod 1.5 XS 2M - 2M | |
HP Z2 Mini G1a | |
BOSGAME M5 | |
Minisforum AI X1 Pro | |
GMK EVO-X2 | |
GMK EVO-T1 | |
Geekom Mini IT15, Ultra 9 285H | |
Minisforum G7 PT | |
Minisforum AtomMan G7 Ti |
Affinity Photo 2 - Combined (Multi CPU) | |
HP Z2 Mini G1a | |
BOSGAME M5 | |
GMK EVO-X2 | |
Minisforum G7 PT | |
Minisforum AtomMan G7 Ti | |
GMK EVO-T1 | |
Minisforum AI X1 Pro | |
Geekom Mini IT15, Ultra 9 285H |
* ... smaller is better
AIDA64: FP32 Ray-Trace | FPU Julia | CPU SHA3 | CPU Queen | FPU SinJulia | FPU Mandel | CPU AES | CPU ZLib | FP64 Ray-Trace | CPU PhotoWorxx
Performance rating | |
BOSGAME M5 | |
GMK EVO-X2 | |
HP Z2 Mini G1a | |
Minisforum G7 PT | |
Minisforum AI X1 Pro | |
Minisforum AtomMan G7 Ti | |
GMK EVO-T1 | |
Average of class Mini PC | |
Geekom Mini IT15, Ultra 9 285H |
AIDA64 / FP32 Ray-Trace | |
BOSGAME M5 | |
GMK EVO-X2 | |
HP Z2 Mini G1a | |
Minisforum G7 PT | |
Minisforum AI X1 Pro | |
GMK EVO-T1 | |
Minisforum AtomMan G7 Ti | |
Average of class Mini PC (770 - 82725, n=111, last 2 years) | |
Geekom Mini IT15, Ultra 9 285H |
AIDA64 / FPU Julia | |
BOSGAME M5 | |
GMK EVO-X2 | |
Minisforum G7 PT | |
HP Z2 Mini G1a | |
Minisforum AI X1 Pro | |
Minisforum AtomMan G7 Ti | |
GMK EVO-T1 | |
Average of class Mini PC (4141 - 227429, n=111, last 2 years) | |
Geekom Mini IT15, Ultra 9 285H |
AIDA64 / CPU SHA3 | |
BOSGAME M5 | |
Minisforum G7 PT | |
GMK EVO-X2 | |
HP Z2 Mini G1a | |
Minisforum AtomMan G7 Ti | |
Minisforum AI X1 Pro | |
GMK EVO-T1 | |
Geekom Mini IT15, Ultra 9 285H | |
Average of class Mini PC (290 - 9475, n=111, last 2 years) |
AIDA64 / CPU Queen | |
Minisforum G7 PT | |
BOSGAME M5 | |
GMK EVO-X2 | |
HP Z2 Mini G1a | |
Minisforum AtomMan G7 Ti | |
Minisforum AI X1 Pro | |
Average of class Mini PC (8208 - 184747, n=111, last 2 years) | |
Geekom Mini IT15, Ultra 9 285H | |
GMK EVO-T1 |
AIDA64 / FPU SinJulia | |
BOSGAME M5 | |
GMK EVO-X2 | |
HP Z2 Mini G1a | |
Minisforum G7 PT | |
Minisforum AI X1 Pro | |
Minisforum AtomMan G7 Ti | |
Average of class Mini PC (408 - 31931, n=111, last 2 years) | |
GMK EVO-T1 | |
Geekom Mini IT15, Ultra 9 285H |
AIDA64 / FPU Mandel | |
BOSGAME M5 | |
GMK EVO-X2 | |
Minisforum G7 PT | |
HP Z2 Mini G1a | |
Minisforum AI X1 Pro | |
GMK EVO-T1 | |
Minisforum AtomMan G7 Ti | |
Average of class Mini PC (2089 - 121827, n=111, last 2 years) | |
Geekom Mini IT15, Ultra 9 285H |
AIDA64 / CPU AES | |
Minisforum G7 PT | |
Minisforum AtomMan G7 Ti | |
BOSGAME M5 | |
GMK EVO-X2 | |
HP Z2 Mini G1a | |
Geekom Mini IT15, Ultra 9 285H | |
Average of class Mini PC (7948 - 325960, n=110, last 2 years) | |
Minisforum AI X1 Pro | |
GMK EVO-T1 |
AIDA64 / CPU ZLib | |
BOSGAME M5 | |
GMK EVO-X2 | |
HP Z2 Mini G1a | |
Minisforum G7 PT | |
Minisforum AtomMan G7 Ti | |
Minisforum AI X1 Pro | |
GMK EVO-T1 | |
Geekom Mini IT15, Ultra 9 285H | |
Average of class Mini PC (127.2 - 2431, n=111, last 2 years) |
AIDA64 / FP64 Ray-Trace | |
GMK EVO-X2 | |
BOSGAME M5 | |
HP Z2 Mini G1a | |
Minisforum G7 PT | |
Minisforum AI X1 Pro | |
GMK EVO-T1 | |
Minisforum AtomMan G7 Ti | |
Average of class Mini PC (432 - 43587, n=111, last 2 years) | |
Geekom Mini IT15, Ultra 9 285H |
AIDA64 / CPU PhotoWorxx | |
BOSGAME M5 | |
GMK EVO-X2 | |
HP Z2 Mini G1a | |
GMK EVO-T1 | |
Minisforum AI X1 Pro | |
Geekom Mini IT15, Ultra 9 285H | |
Minisforum AtomMan G7 Ti | |
Minisforum G7 PT | |
Average of class Mini PC (5333 - 94259, n=111, last 2 years) |
Cinebench R15 Multi endurance test
System performance
The GMKtec EVO-X2 offers excellent system performance. Its strengths are particularly evident in the RAM benchmarks. The integrated NPU with 50 TOPS accelerates AI applications and is on a par with the Minisforum AI X1 Pro in terms of performance.
WebXPRT 4: Overall
WebXPRT 3: Overall
Mozilla Kraken 1.1: Total
CrossMark: Overall | Productivity | Creativity | Responsiveness
* ... smaller is better
AIDA64 / Memory Copy | |
HP Z2 Mini G1a | |
GMK EVO-X2 | |
BOSGAME M5 | |
GMK EVO-T1 | |
Geekom Mini IT15, Ultra 9 285H | |
Minisforum AI X1 Pro | |
Average of class Mini PC (8472 - 163031, n=111, last 2 years) |
AIDA64 / Memory Read | |
GMK EVO-X2 | |
BOSGAME M5 | |
HP Z2 Mini G1a | |
Minisforum AI X1 Pro | |
Geekom Mini IT15, Ultra 9 285H | |
GMK EVO-T1 | |
Average of class Mini PC (9130 - 123769, n=111, last 2 years) |
AIDA64 / Memory Write | |
HP Z2 Mini G1a | |
BOSGAME M5 | |
GMK EVO-X2 | |
Minisforum AI X1 Pro | |
Geekom Mini IT15, Ultra 9 285H | |
GMK EVO-T1 | |
Average of class Mini PC (9047 - 216693, n=111, last 2 years) |
AIDA64 / Memory Latency | |
HP Z2 Mini G1a | |
BOSGAME M5 | |
GMK EVO-X2 | |
GMK EVO-T1 | |
Geekom Mini IT15, Ultra 9 285H | |
Minisforum AI X1 Pro | |
Average of class Mini PC (82.3 - 184.8, n=111, last 2 years) |
* ... smaller is better
NPU Performance
UL Procyon for Windows / Overall Score Integer NPU | |
BOSGAME M5 (AMD Ryzen AI) | |
GMK EVO-X2 (AMD Ryzen AI) | |
HP Z2 Mini G1a (AMD Ryzen AI) | |
Minisforum AI X1 Pro (AMD Ryzen AI) | |
GMK EVO-T1 (OpenVINO) | |
Geekom Mini IT15, Ultra 9 285H (OpenVINO) |
UL Procyon for Windows / Overall Score Integer GPU | |
BOSGAME M5 (Windows ML) | |
GMK EVO-X2 (Windows ML) | |
HP Z2 Mini G1a (Windows ML) | |
Minisforum AI X1 Pro (Windows ML) | |
GMK EVO-T1 (Windows ML) | |
Geekom Mini IT15, Ultra 9 285H (Windows ML) |
DPC latencies
The GMKtec EVO-X2 delivers solid DPC latencies in the test and is only in the midfield in comparison. According to LatencyMon, however, the latencies are slightly higher. There were no problems with 4K video playback.
DPC Latencies / LatencyMon - interrupt to process latency (max), Web, Youtube, Prime95 | |
Minisforum AI X1 Pro | |
HP Z2 Mini G1a | |
GMK EVO-X2 | |
BOSGAME M5 | |
Minisforum G7 PT | |
GMK EVO-T1 | |
Geekom Mini IT15, Ultra 9 285H |
* ... smaller is better
Storage devices
Drive Performance rating - Percent | |
Geekom Mini IT15, Ultra 9 285H | |
GMK EVO-T1 | |
GMK EVO-X2 | |
HP Z2 Mini G1a | |
BOSGAME M5 | |
Minisforum AI X1 Pro | |
Average of class Mini PC |
* ... smaller is better
Sustained read load: DiskSpd Read Loop, Queue Depth 8
Graphics card and gaming performance
The powerful AMD Radeon RX 8060S of the GMKtec EVO-X2 achieves graphics performance at the level of an Nvidia GeForce RTX 4070 laptop. This makes the GMKtec EVO-X2 as well as the Bosgame and the HP Z2 Mini G1a among the most powerful systems with iGPU in the consumer sector. Current games run smoothly, provided they are mainly played in 1,080p. The iGPU can also handle older games in 1,440p or 2,160p
3DMark Performance rating - Percent | |
HP Z2 Mini G1a | |
BOSGAME M5 | |
Minisforum AtomMan G7 Ti | |
GMK EVO-X2 | |
Minisforum G7 PT | |
GMK EVO-T1 | |
Minisforum AI X1 Pro | |
Geekom Mini IT15, Ultra 9 285H |
3DMark 11 - 1280x720 Performance GPU | |
HP Z2 Mini G1a | |
BOSGAME M5 | |
GMK EVO-X2 | |
Minisforum G7 PT | |
Minisforum AtomMan G7 Ti | |
Minisforum AI X1 Pro | |
Geekom Mini IT15, Ultra 9 285H | |
GMK EVO-T1 |
Unigine Valley 1.0 | |
1920x1080 Extreme HD DirectX AA:x8 | |
Minisforum AtomMan G7 Ti | |
GMK EVO-X2 | |
BOSGAME M5 | |
HP Z2 Mini G1a | |
Minisforum G7 PT | |
Minisforum AI X1 Pro | |
GMK EVO-T1 | |
Geekom Mini IT15, Ultra 9 285H | |
1920x1080 Extreme HD Preset OpenGL AA:x8 | |
Minisforum AtomMan G7 Ti | |
GMK EVO-X2 | |
BOSGAME M5 | |
HP Z2 Mini G1a | |
Minisforum G7 PT | |
GMK EVO-T1 | |
Geekom Mini IT15, Ultra 9 285H | |
Minisforum AI X1 Pro |
Affinity Photo 2 - Combined (Single / Multi GPU) | |
Minisforum G7 PT | |
GMK EVO-T1 | |
Geekom Mini IT15, Ultra 9 285H | |
Minisforum AtomMan G7 Ti | |
BOSGAME M5 | |
GMK EVO-X2 | |
HP Z2 Mini G1a | |
Minisforum AI X1 Pro |
SPECworkstation 4.0: Graphics Score | Accelerator Score | Viewport Graphics | HandBrake | ONNX Inference
Performance rating | |
BOSGAME M5 | |
GMK EVO-X2 | |
HP Z2 Mini G1a | |
Minisforum G7 PT | |
Minisforum AtomMan G7 Ti | |
Average NVIDIA GeForce RTX 4050 Laptop GPU | |
Minisforum AI X1 Pro | |
GMK EVO-T1 | |
Geekom Mini IT15, Ultra 9 285H |
SPECworkstation 4.0 / Graphics Score | |
BOSGAME M5 | |
GMK EVO-X2 | |
HP Z2 Mini G1a | |
Minisforum G7 PT | |
Minisforum AtomMan G7 Ti | |
Average NVIDIA GeForce RTX 4050 Laptop GPU (n=1) | |
Minisforum AI X1 Pro | |
Geekom Mini IT15, Ultra 9 285H | |
GMK EVO-T1 |
SPECworkstation 4.0 / Accelerator Score | |
BOSGAME M5 | |
GMK EVO-X2 | |
HP Z2 Mini G1a | |
Minisforum AtomMan G7 Ti | |
Minisforum G7 PT | |
Average NVIDIA GeForce RTX 4050 Laptop GPU (n=1) | |
Minisforum AI X1 Pro | |
Geekom Mini IT15, Ultra 9 285H | |
GMK EVO-T1 |
SPECworkstation 4.0 / Viewport Graphics | |
BOSGAME M5 | |
GMK EVO-X2 | |
HP Z2 Mini G1a | |
Minisforum G7 PT | |
Minisforum AtomMan G7 Ti | |
Average NVIDIA GeForce RTX 4050 Laptop GPU (n=1) | |
Minisforum AI X1 Pro | |
Geekom Mini IT15, Ultra 9 285H | |
GMK EVO-T1 |
SPECworkstation 4.0 / HandBrake | |
BOSGAME M5 | |
Minisforum G7 PT | |
GMK EVO-X2 | |
HP Z2 Mini G1a | |
Minisforum AI X1 Pro | |
Minisforum AtomMan G7 Ti | |
GMK EVO-T1 | |
Geekom Mini IT15, Ultra 9 285H | |
Average NVIDIA GeForce RTX 4050 Laptop GPU (n=1) |
SPECworkstation 4.0 / ONNX Inference | |
HP Z2 Mini G1a | |
BOSGAME M5 | |
GMK EVO-X2 | |
Minisforum G7 PT | |
Minisforum AtomMan G7 Ti | |
Minisforum AI X1 Pro | |
GMK EVO-T1 | |
Average NVIDIA GeForce RTX 4050 Laptop GPU (n=1) | |
Geekom Mini IT15, Ultra 9 285H |
GMK EVO-X2, AMD Radeon 8060S Ryzen AI Max+ 395 | BOSGAME M5, AMD Radeon 8060S Ryzen AI Max+ 395 | HP Z2 Mini G1a, AMD Radeon 8060S Ryzen AI Max+ PRO 395 | Minisforum AtomMan G7 Ti, NVIDIA GeForce RTX 4070 Laptop GPU i9-14900HX | Minisforum G7 PT, AMD Radeon RX 7600M XT R9 7945HX | Minisforum AI X1 Pro, AMD Radeon 890M Ryzen AI 9 HX 370 | Geekom Mini IT15, Ultra 9 285H, Intel Arc Graphics 140T Ultra 9 285H | GMK EVO-T1, Intel Arc Graphics 140T Ultra 9 285H | Average of class Mini PC | |
---|---|---|---|---|---|---|---|---|---|
SPECviewperf 2020 | -1% | -4% | -37% | -21% | -72% | -83% | -85% | -64% | |
3840x2160 Medical (medical-03) (fps) | 89.7 | 90.1 0% | 87.5 -2% | 15.2 -83% | 30.8 -66% | 15.2 -83% | 7.08 -92% | 7.02 -92% | 17.2 ? -81% |
3840x2160 Energy (energy-03) (fps) | 75.7 | 76 0% | 73.7 -3% | 44 -42% | 43.2 -43% | 15.5 -80% | 7.94 -90% | 7.4 -90% | 23.9 ? -68% |
3840x2160 Creo (creo-03) (fps) | 97.3 | 96.2 -1% | 90.9 -7% | 93.7 -4% | 97.5 0% | 34.1 -65% | 26.7 -73% | 19.9 -80% | 41.2 ? -58% |
3840x2160 CATIA (catia-06) (fps) | 70.3 | 69.7 -1% | 68.1 -3% | 37.3 -47% | 64.1 -9% | 23.1 -67% | 14.6 -79% | 12.8 -82% | 27.5 ? -61% |
3840x2160 3ds Max (3dsmax-07) (fps) | 73.5 | 73.6 0% | 71.2 -3% | 69 -6% | 63.9 -13% | 21.9 -70% | 17.8 -76% | 14.6 -80% | 36.7 ? -50% |
3840x2160 Siemens NX (snx-04) (fps) | 233 | 231 -1% | 221 -5% | 19.4 -92% | 220 -6% | 68.7 -71% | 6.81 -97% | 6.13 -97% | 50.5 ? -78% |
3840x2160 Maya (maya-06) (fps) | 196 | 194 -1% | 186 -5% | 223 14% | 171 -13% | 55.9 -71% | 51.3 -74% | 44.9 -77% | 93.5 ? -52% |
3DMark 05 Standard | 79090 points | |
3DMark 06 Standard Score | 65432 points | |
3DMark Vantage P Result | 98782 points | |
3DMark 11 Performance | 34534 points | |
3DMark Ice Storm Standard Score | 249213 points | |
3DMark Cloud Gate Standard Score | 69003 points | |
3DMark Fire Strike Score | 26889 points | |
3DMark Fire Strike Extreme Score | 13420 points | |
3DMark Time Spy Score | 11094 points | |
3DMark Steel Nomad Score | 2196 points | |
Help |
GTA V | |
1024x768 Lowest Settings possible | |
Minisforum G7 PT | |
Minisforum AtomMan G7 Ti | |
BOSGAME M5 | |
GMK EVO-X2 | |
HP Z2 Mini G1a | |
Geekom Mini IT15, Ultra 9 285H | |
Minisforum AI X1 Pro | |
GMK EVO-T1 | |
1920x1080 High/On (Advanced Graphics Off) AA:2xMSAA + FX AF:8x | |
Minisforum G7 PT | |
Minisforum AtomMan G7 Ti | |
BOSGAME M5 | |
HP Z2 Mini G1a | |
GMK EVO-X2 | |
Minisforum AI X1 Pro | |
GMK EVO-T1 | |
Geekom Mini IT15, Ultra 9 285H |
Cyberpunk 2077 2.2 Phantom Liberty | |
1920x1080 Low Preset (FSR off) | |
HP Z2 Mini G1a | |
BOSGAME M5 | |
GMK EVO-X2 | |
Minisforum AtomMan G7 Ti | |
Minisforum G7 PT | |
Minisforum AI X1 Pro | |
GMK EVO-T1 | |
Geekom Mini IT15, Ultra 9 285H | |
1920x1080 High Preset (FSR off) | |
HP Z2 Mini G1a | |
BOSGAME M5 | |
GMK EVO-X2 | |
Minisforum AtomMan G7 Ti | |
Minisforum G7 PT | |
Minisforum AI X1 Pro | |
GMK EVO-T1 | |
Geekom Mini IT15, Ultra 9 285H |
Baldur's Gate 3 | |
1280x720 Low Preset AA:SM | |
Minisforum G7 PT | |
BOSGAME M5 | |
GMK EVO-X2 | |
HP Z2 Mini G1a | |
Minisforum AtomMan G7 Ti | |
Minisforum AI X1 Pro | |
Geekom Mini IT15, Ultra 9 285H | |
GMK EVO-T1 | |
1920x1080 High Preset AA:T | |
Minisforum AtomMan G7 Ti | |
BOSGAME M5 | |
HP Z2 Mini G1a | |
GMK EVO-X2 | |
Minisforum G7 PT | |
Minisforum AI X1 Pro | |
GMK EVO-T1 | |
Geekom Mini IT15, Ultra 9 285H |
F1 24 | |
1920x1080 Low Preset AA:T AF:16x | |
Minisforum G7 PT | |
BOSGAME M5 | |
Minisforum AtomMan G7 Ti | |
HP Z2 Mini G1a | |
GMK EVO-X2 | |
GMK EVO-T1 | |
Minisforum AI X1 Pro | |
Geekom Mini IT15, Ultra 9 285H | |
1920x1080 High Preset AA:T AF:16x | |
Minisforum G7 PT | |
BOSGAME M5 | |
GMK EVO-X2 | |
Minisforum AtomMan G7 Ti | |
HP Z2 Mini G1a | |
GMK EVO-T1 | |
Geekom Mini IT15, Ultra 9 285H | |
Minisforum AI X1 Pro |
30 min. Cyberpunk 2077 fps chart (1920 x 1080, Ultra preset, FSR off)
low | med. | high | ultra | QHD FSR | QHD | 4K FSR | 4K | |
---|---|---|---|---|---|---|---|---|
GTA V (2015) | 184.2 | 181.7 | 168.6 | 100.9 | 77.2 | 80.3 | ||
The Witcher 3 (2015) | 525 | 353 | 193.9 | 112.3 | 84.6 | 59.6 | ||
Dota 2 Reborn (2015) | 217.5 | 185.2 | 184.7 | 138.9 | 98.2 | |||
Final Fantasy XV Benchmark (2018) | 204 | 137 | 96.7 | 70.6 | 38.7 | |||
X-Plane 11.11 (2018) | 129.5 | 98.8 | 76.5 | 59.5 | ||||
Strange Brigade (2018) | 608 | 280 | 231 | 199.1 | 123.8 | 62.4 | ||
Anno 1800 (2019) | 176.9 | 143.6 | 107.5 | 55.6 | 49.5 | 37 | ||
The Witcher 3 v4 (2023) | 194.1 | 164.4 | 141.2 | 120 | 76.6 | 37.7 | ||
Baldur's Gate 3 (2023) | 152.3 | 120.2 | 100.6 | 100.4 | 68.4 | 53.9 | 32.6 | |
Cyberpunk 2077 2.2 Phantom Liberty (2023) | 148.1 | 125.1 | 99.7 | 89.6 | 54.5 | 22.2 | ||
Avatar Frontiers of Pandora (2023) | 76 | 69 | 59 | 45 | 30 | 15 | ||
F1 24 (2024) | 265 | 243 | 193.1 | 58.5 | 62.7 | 38.7 | 34.2 | 19.5 |
Black Myth: Wukong (2024) | 93 | 69 | 53 | 29 | 22 | 12 | ||
Star Wars Outlaws (2024) | 55.6 | 50.3 | 44.2 | 36.3 | 37.8 | 26.2 | 23.5 | 14.7 |
Space Marine 2 (2024) | 80.9 | 74.9 | 67.5 | 46.6 | 60.5 | 26.2 | 39.5 | |
Indiana Jones and the Great Circle (2024) | 92.3 | 65.8 | 59.8 | 54.7 | 42 | 23.3 | ||
Monster Hunter Wilds (2025) | 72.3 | 62 | 47.4 | 34 | 18.6 | |||
Assassin's Creed Shadows (2025) | 72 | 68 | 57 | 40 | 23 | 34 | 33 | 22 |
F1 25 (2025) | 235 | 204 | 163.1 |
Emissions
Noise emissions
In the volume measurements, the GMKtec EVO-X2 is audible under full load, but remains significantly quieter than the HP Z2 Mini G1a. In quiet mode, the background noise is noticeably reduced to around 41 dB(A), although this is accompanied by lower performance. At a fixed fan speed of 20 percent, we only measured 37 dB(A).
Noise level
Idle |
| 27.4 / 27.5 / 27.5 dB(A) |
Load |
| 47.5 / 47.2 dB(A) |
![]() | ||
30 dB silent 40 dB(A) audible 50 dB(A) loud |
||
min: ![]() ![]() ![]() |
GMK EVO-X2 Ryzen AI Max+ 395, Radeon 8060S | BOSGAME M5 Ryzen AI Max+ 395, Radeon 8060S | HP Z2 Mini G1a Ryzen AI Max+ PRO 395, Radeon 8060S | Minisforum AtomMan G7 Ti i9-14900HX, NVIDIA GeForce RTX 4070 Laptop GPU | Minisforum G7 PT R9 7945HX, Radeon RX 7600M XT | Minisforum AI X1 Pro Ryzen AI 9 HX 370, Radeon 890M | Geekom Mini IT15, Ultra 9 285H Ultra 9 285H, Arc 140T | GMK EVO-T1 Ultra 9 285H, Arc 140T | |
---|---|---|---|---|---|---|---|---|
Noise | 3% | -22% | -13% | -0% | 7% | -10% | -17% | |
off / environment * (dB) | 24.9 | 24.8 -0% | 24.9 -0% | 24.7 1% | 24.4 2% | 23.8 4% | 24.3 2% | 24.7 1% |
Idle Minimum * (dB) | 27.4 | 25 9% | 32.1 -17% | 34 -24% | 29.3 -7% | 28 -2% | 33.7 -23% | 38.7 -41% |
Idle Average * (dB) | 27.5 | 25.1 9% | 32.4 -18% | 34.1 -24% | 29.4 -7% | 28.1 -2% | 33.7 -23% | 38.8 -41% |
Idle Maximum * (dB) | 27.5 | 25.1 9% | 32.6 -19% | 34.1 -24% | 29.4 -7% | 28.2 -3% | 33.8 -23% | 38.9 -41% |
Load Average * (dB) | 47.5 | 48.8 -3% | 41.1 13% | 48.6 -2% | 43.5 8% | 38.4 19% | 47.6 -0% | 51.6 -9% |
Witcher 3 ultra * (dB) | 47.4 | 46.5 2% | 70 -48% | 38.7 18% | ||||
Cyberpunk 2077 ultra * (dB) | 47.3 | 46.6 1% | 67 -42% | 55.2 -17% | 45.5 4% | 38.5 19% | 47.7 -1% | 52 -10% |
Load Maximum * (dB) | 47.2 | 48.6 -3% | 67.3 -43% | 48.3 -2% | 45 5% | 41.8 11% | 47.9 -1% | 51.6 -9% |
* ... smaller is better
Temperature
Under load, the housing of the GMKtec EVO-X2 only heats up slightly. The metal surfaces ensure good heat distribution so that no hotspots occur. The processor works as intended with up to 140 watts for short periods, but stabilizes at 120 watts under prolonged load.
(+) The maximum temperature on the upper side is 32.5 °C / 91 F, compared to the average of 35.8 °C / 96 F, ranging from 21.6 to 69.5 °C for the class Mini PC.
(+) The bottom heats up to a maximum of 37.1 °C / 99 F, compared to the average of 36.5 °C / 98 F
(+) In idle usage, the average temperature for the upper side is 25.1 °C / 77 F, compared to the device average of 32.4 °C / 90 F.
(±) 3: The average temperature for the upper side is 34.2 °C / 94 F, compared to the average of 32.4 °C / 90 F for the class Mini PC.
Energy management
The GMKtec EVO-X2 impresses with its low power consumption in relation to the performance it offers. In idle mode, the EVO-X2 only requires 6.5 to 10.5 watts. In terms of efficiency, our test sample only reached the midfield. Compared to devices with a dedicated GPU, the EVO-X2 can still hold its own.
Off / Standby | ![]() ![]() |
Idle | ![]() ![]() ![]() |
Load |
![]() ![]() |
![]() ![]() ![]() ![]() ![]() ![]() ![]() | |
Key:
min: ![]() ![]() ![]() |
GMK EVO-X2 Ryzen AI Max+ 395, Radeon 8060S | BOSGAME M5 Ryzen AI Max+ 395, Radeon 8060S | HP Z2 Mini G1a Ryzen AI Max+ PRO 395, Radeon 8060S | Minisforum G7 PT R9 7945HX, Radeon RX 7600M XT | Minisforum AtomMan G7 Ti i9-14900HX, NVIDIA GeForce RTX 4070 Laptop GPU | Minisforum AI X1 Pro Ryzen AI 9 HX 370, Radeon 890M | Geekom Mini IT15, Ultra 9 285H Ultra 9 285H, Arc 140T | GMK EVO-T1 Ultra 9 285H, Arc 140T | Average AMD Radeon 8060S | Average of class Mini PC | |
---|---|---|---|---|---|---|---|---|---|---|
Power Consumption | -22% | -12% | -156% | -149% | 15% | 45% | 20% | -13% | -11% | |
Idle Minimum * (Watt) | 6.5 | 9.3 -43% | 7.1 -9% | 23 -254% | 21 -223% | 8.6 -32% | 3.8 42% | 6.1 6% | 8.2 ? -26% | 9.49 ? -46% |
Idle Average * (Watt) | 8.2 | 11.5 -40% | 9.5 -16% | 28.1 -243% | 35.1 -328% | 9.4 -15% | 5.2 37% | 7.9 4% | 10.4 ? -27% | 11.6 ? -41% |
Idle Maximum * (Watt) | 10.5 | 16.4 -56% | 13.8 -31% | 40.4 -285% | 39.6 -277% | 10 5% | 8.4 20% | 13.2 -26% | 13.4 ? -28% | 15.1 ? -44% |
Load Average * (Watt) | 84.3 | 83.9 -0% | 76.7 9% | 131.2 -56% | 106.9 -27% | 70.2 17% | 50.1 41% | 60.7 28% | 82.9 ? 2% | 63 ? 25% |
Cyberpunk 2077 ultra external monitor * (Watt) | 178.1 | 174.2 2% | 195.1 -10% | 282 -58% | 217 -22% | 70.7 60% | 51.6 71% | 67.4 62% | ||
Load Maximum * (Watt) | 214.9 | 200.9 7% | 245.4 -14% | 296 -38% | 249 -16% | 97.7 55% | 91.9 57% | 117.1 46% | 184.8 ? 14% | 108.4 ? 50% |
* ... smaller is better
Power efficiency
Performance rating - Percent | |
Geekom Mini IT15, Ultra 9 285H | |
Minisforum AI X1 Pro | |
GMK EVO-T1 | |
BOSGAME M5 | |
GMK EVO-X2 | |
Minisforum G7 PT | |
HP Z2 Mini G1a | |
Minisforum AtomMan G7 Ti |
Power Consumption | |
Cinebench R23 Multi Power Efficiency - external Monitor | |
Minisforum AI X1 Pro | |
Geekom Mini IT15, Ultra 9 285H | |
Minisforum G7 PT | |
GMK EVO-T1 | |
BOSGAME M5 | |
GMK EVO-X2 | |
HP Z2 Mini G1a | |
Minisforum AtomMan G7 Ti | |
Cinebench R23 Single Power Efficiency - external Monitor | |
Geekom Mini IT15, Ultra 9 285H | |
GMK EVO-T1 | |
Minisforum AI X1 Pro | |
HP Z2 Mini G1a | |
GMK EVO-X2 | |
BOSGAME M5 | |
Minisforum G7 PT | |
Minisforum AtomMan G7 Ti | |
Cinebench R15 Multi Efficiency (external Monitor) | |
Minisforum AI X1 Pro | |
Geekom Mini IT15, Ultra 9 285H | |
Minisforum G7 PT | |
BOSGAME M5 | |
GMK EVO-T1 | |
GMK EVO-X2 | |
HP Z2 Mini G1a | |
Minisforum AtomMan G7 Ti |
Power consumption - external monitor
Power consumption - external monitor
Notebookcheck rating
GMK EVO-X2
- 08/28/2025 v8
Sebastian Bade
Possible alternatives in comparison
Image | Model / Review | Price | Weight | Dimensions |
---|---|---|---|---|
GMK EVO-X2 AMD Ryzen AI Max+ 395 ⎘ AMD Radeon 8060S ⎘ 64 GB Memory, 1024 GB SSD | Amazon: $1,999.99 List Price: 1499€ | 1.6 kg | 185.8 x 193 x 77 mm | |
BOSGAME M5 AMD Ryzen AI Max+ 395 ⎘ AMD Radeon 8060S ⎘ 128 GB Memory, 2024 GB SSD | Amazon: 1. $23.99 19V 3A 3.42A MINI PC Power A... 2. $21.99 UL Listed 19V 3A 3.42A AC Ad... 3. $23.65 Mini PC Computer Protective ... List Price: 1469€ | 1.3 kg | 200 x 215 x 57 mm | |
Minisforum G7 PT AMD Ryzen 9 7945HX ⎘ AMD Radeon RX 7600M XT ⎘ 32 GB Memory, 1024 GB SSD | Amazon: List Price: 1359€ | 2.8 kg | 220 x 155 x 285 mm | |
Minisforum AI X1 Pro AMD Ryzen AI 9 HX 370 ⎘ AMD Radeon 890M ⎘ 64 GB Memory, 1024 GB SSD | Amazon: 1. $1,119.90 MINISFORUM AI X1 Pro-370 Min... 2. $999.00 MINISFORUM Mini PC AI X1 Pro... 3. $1,019.00 MINISFORUM N5 Pro-P370 5-Bay... | 1.4 kg | 195 x 195 x 42.5 mm | |
Minisforum AtomMan G7 Ti Intel Core i9-14900HX ⎘ NVIDIA GeForce RTX 4070 Laptop GPU ⎘ 32 GB Memory, 1024 GB SSD | Amazon: 1. $39.00 New Replacement MINI PC CPU ... 2. $199.99 HIGOLEPC Mini PC Win 11 Pro,... 3. $369.99 MINISFORUM UM760 Slim Mini P... List Price: 2049 EUR | 2.6 kg | 260 x 80 x 395 mm | |
HP Z2 Mini G1a AMD Ryzen AI Max+ PRO 395 ⎘ AMD Radeon 8060S ⎘ 64 GB Memory, 2048 GB SSD | Amazon: 1. $2,613.56 HP Z2 Mini G1a Workstation -... 2. $1,924.43 HP Z2 Mini G1a Workstation -... 3. $2,099.99 HP Z2 Mini G1a Workstation -... List Price: 3175 EUR | 2.6 kg | 200 x 168 x 85.5 mm | |
Geekom Mini IT15, Ultra 9 285H Intel Core Ultra 9 285H ⎘ Intel Arc Graphics 140T ⎘ 32 GB Memory, 2048 GB SSD | Amazon: $1,199.00 List Price: 1099€ | 595 g | 112 x 117 x 49.2 mm | |
GMK EVO-T1 Intel Core Ultra 9 285H ⎘ Intel Arc Graphics 140T ⎘ 64 GB Memory, 1024 GB SSD | Amazon: $1,299.99 List Price: 999€ | 946 g | 151 x 154 x 73.6 mm |
Transparency
The selection of devices to be reviewed is made by our editorial team. The test sample was given to the author by the manufacturer free of charge for the purposes of review. There was no third-party influence on this review, nor did the manufacturer receive a copy of this review before publication. There was no obligation to publish this review. As an independent media company, Notebookcheck is not subjected to the authority of manufacturers, retailers or publishers.
This is how Notebookcheck is testing
Every year, Notebookcheck independently reviews hundreds of laptops and smartphones using standardized procedures to ensure that all results are comparable. We have continuously developed our test methods for around 20 years and set industry standards in the process. In our test labs, high-quality measuring equipment is utilized by experienced technicians and editors. These tests involve a multi-stage validation process. Our complex rating system is based on hundreds of well-founded measurements and benchmarks, which maintains objectivity. Further information on our test methods can be found here.