
BOSGAME M5 vs GMKtec EVO-X2
Two mini PCs side by side: what each one suits, where they differ, and the price of each on amazon.com.
Updated


GMKtec EVO-X2
The short answer
Both machines run the same AMD Ryzen AI Max+ 395 with Radeon 8060S graphics, so the choice comes down to configurations and outputs. The BOSGAME M5 comes one way, 128 GB of LPDDR5X with 2000 GB of storage, and suits a buyer who already knows they want the large-memory setup. The GMKtec EVO-X2 suits someone who wants to pick a memory and storage size, or who needs dedicated HDMI and DisplayPort outputs alongside USB-C. GMKtec also documents more of its box, from the dimensions to the power supply.
Specifications side by side
CPU, GPU, installed memory and storage describe the configurations shown above. Memory limits, slots and ports describe the chassis; other configurations can differ.
| Specification | BOSGAME M5 | GMKtec EVO-X2 |
|---|---|---|
| Processor | AMD Ryzen AI Max+ 395 | AMD Ryzen AI Max+ 395 |
| Graphics | AMD Radeon 8060S | AMD Radeon 8060S |
| Installed memory | 128 GB | 64 GB |
| Installed storage | 2 TB | 1 TB |
| Maximum memory | 128 GB | 128 GB |
| Memory slots | 0 | 0 |
| M.2 slots | 2 | 2 |
| LAN speed | 2.5GbE | 2.5GbE |
| USB4 ports | 2 | 2 |
| Thunderbolt ports | 0 | 0 |
Workload by workload
For everyday work and coding, both are well suited, and the reason is shared: 16 cores and 32 threads on each. Gaming is workable on both, because the Radeon 8060S is integrated and neither has a discrete GPU.
Local AI is well suited on both, and that rests on memory capacity. For local AI, check the exact model release, quantization, context length, runtime overhead and memory available after the operating system. System RAM and GPU VRAM are different limits; shared-memory allocation depends on firmware, drivers and the backend. CPU inference or partial offloading can change what loads and its speed. Installed capacity alone does not establish compatibility or useful generation speed. Check the current Ollama or LM Studio hardware documentation for the exact GPU and operating system. Generation speed depends on memory bandwidth, which neither maker states, though both quote 8000 MT/s memory.
Workstation use is well suited on the BOSGAME M5 with its 128 GB. The EVO-X2 is well suited in its 64 GB LPDDR5X version and both 128 GB versions, but only workable in the 64 GB DDR5 version with 2000 GB of storage. If workstation work is the reason you are shopping, skip that one.
Homelab use is workable on both, for the same reason: a single 2.5GbE port limits routing or network separation on the machine itself, and the memory is soldered.
Expansion and upgrades
Memory is fixed on both: there are no memory slots, it is soldered, and the maximum is 128 GB. The BOSGAME M5 already sits at that maximum. A 64 GB EVO-X2 sits 64 GB below it with no way to close the gap, so choose the memory size at purchase.
Storage is the one upgrade path. Each has two M.2 slots for 2280 drives, the first at PCIe 4.0. An added drive in the other slot is an M.2 drive whose speed depends on that slot's generation, because a drive runs at the speed of its slot. The EVO-X2 has no SATA bays. Neither has a PCIe slot or OCuLink.
Ports and connectivity
Much is shared: one 2.5GbE port, Wi-Fi 7 with the wireless module included, two USB4 ports, five USB-A ports and an SD card reader on each, with no Thunderbolt. Both USB-C ports carry DisplayPort, so either machine can drive screens over USB-C.
The EVO-X2 adds one HDMI 2.1 and one DisplayPort 1.4 output. HDMI 2.1 features also need a matching cable and display. BOSGAME does not state HDMI or DisplayPort outputs, and neither maker states the maximum number of displays. The EVO-X2 has Bluetooth 5.4 and a 19.5V 11.8A external (230W) supply.
BOSGAME also does not state dimensions, volume, Bluetooth, power supply, TDP, VESA mounting or ECC support; GMKtec does not state TDP, VESA mounting or ECC support.
Which configurations to compare
The fair pairing is the BOSGAME M5 with 128 GB of LPDDR5X and 2000 GB of storage against the EVO-X2 with 128 GB of LPDDR5X and 2000 GB. Processor, graphics and memory match, so what separates them is the stated display outputs and how fully each maker documents the box.
If a smaller drive is enough, the 128 GB EVO-X2 with 1000 GB keeps the memory and gives up storage. The 64 GB DDR5 version uses a different memory type and is only workable for workstation use.
If neither fits
These two share a lot, including soldered memory, a single network port and no PCIe slot, so if one of those rules them out, it rules out both. The alternative to look at is the GEEKOM A9 Mega, a close match to these machines. Treat it as the next page to read, and confirm on its own specifications whether it removes the particular limit that stopped you here.
Workload fit
The best configuration of each machine, as our fit rules judge it.
| Workload | BOSGAME M5 | GMKtec EVO-X2 |
|---|---|---|
| Everyday | Good | Good |
| Coding | Good | Good |
| Gaming | Workable | Workable |
| Home lab | Workable | Workable |
| Local AI | Good | Good |
| Workstation | Good | Good |
Common questions
Can I add more memory to either machine later?
No. Both have soldered memory, no memory slots and a 128 GB maximum, so the size you buy is the size you keep. If you might want 128 GB later, buy it now.
Which versions can run a 120B language model?
For local AI, check the exact model release, quantization, context length, runtime overhead and memory available after the operating system. System RAM and GPU VRAM are different limits; shared-memory allocation depends on firmware, drivers and the backend. CPU inference or partial offloading can change what loads and its speed. Installed capacity alone does not establish compatibility or useful generation speed. Check the current Ollama or LM Studio hardware documentation for the exact GPU and operating system.
How many monitors can I connect?
Neither maker states a maximum. The EVO-X2 has one HDMI 2.1 and one DisplayPort 1.4 output, and both machines carry DisplayPort over USB-C. BOSGAME does not state HDMI or DisplayPort outputs on the M5.
Do they have an NPU for AI work?
Yes, both quote 50 NPU TOPS in every configuration. Their suitability for local AI rests on memory capacity rather than the NPU figure, so pick the memory size first for language models.
The two machines
Fields we could not confirm from the manufacturer are marked not confirmed rather than guessed. Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon.com at the time of purchase will apply to the purchase of this product.