readygpu

GMKtec EVO-X2 Ryzen AI Max+ 395 128GB vs Mac mini M4 32GB

For running models locally, priced in the US. At 64k context the first runs 74 models fully in memory, the second 38.

In short

The GMKtec EVO-X2 Ryzen AI Max+ 395 128GB has more memory for models: 128 GB against 32 GB. At 64k context the GMKtec EVO-X2 Ryzen AI Max+ 395 128GB runs 74 of the models we track fully in memory and the Mac mini M4 32GB 38; only the GMKtec EVO-X2 Ryzen AI Max+ 395 128GB holds Qwen 3.8 Flash Next. On the models both run fully in memory, the GMKtec EVO-X2 Ryzen AI Max+ 395 128GB is about 1.8× as fast as the Mac mini M4 32GB (estimated, median across models). In the US the GMKtec EVO-X2 Ryzen AI Max+ 395 128GB costs $1,600 at Amazon and the Mac mini M4 32GB $1,420 at eBay: the Mac mini M4 32GB is $180 cheaper.

Updated . Speeds are estimates; how they are worked out.

GMKtec EVO-X2 Ryzen AI Max+ 395 128GB

$1,600

Amazon

Add to build
Mac mini M4 32GB

$1,420

eBay

Add to build

Specifications

SpecGMKtec EVO-X2 Ryzen AI Max+ 395 128GBMac mini M4 32GB
ChipRyzen AI Max+ 395 (Radeon 8060S)Apple M4 (10-core GPU)
Memory bus256-bit128-bit
Memory bandwidth256 GB/s120 GB/s
FP16 compute59.4 TFLOPS3.76 TFLOPS
SoftwareROCmMetal
Unified memory128 GB32 GB
GPU can use by default96 GB21.3 GB
GPU can use, tuned120 GB24 GB
Operating systemwindowsmacos
ProcessorRyzen AI Max+ 395M4 10-core
Storage2000 GB512 GB

What each runs

Estimated decode speed with the most precise quant that fits each one est.

ModelGMKtec EVO-X2 Ryzen AI Max+ 395 128GBMac mini M4 32GB
Qwen 3.8 Flash Next3tok/sUD-Q3_K_XLwon't run
MiniMax M2.74tok/sUD-Q2_K_XLwon't run
Qwen 3.8 27B4tok/sQ8_04tok/sUD-Q4_K_M
Qwen 3.5 122B-A10B9tok/sQ5_K_Mwon't run
Qwen 3.6 27B4tok/sQ8_04tok/sQ4_K_M
gpt-oss 120B15tok/sMXFP4won't run
Qwen3 Coder Next15tok/sQ8_0won't run
Nemotron 3 Super 120B-A12B8tok/sQ5_K_Mwon't run
Gemma 4 31B4tok/sQ8_0won't run
Mistral Small 4 119B11tok/sUD-Q6_Kwon't run
Qwen 3.6 35B-A3B19tok/sQ8_0won't run
Qwen 3.5 27B4tok/sQ8_04tok/sQ4_K_M
Gemma 4 26B-A4B15tok/sQ8_010tok/sQ5_K_M
GLM 4.5 Air4tok/sQ5_K_Mwon't run
Qwen 3.5 35B-A3B19tok/sQ8_0won't run
Qwen3 Next 80B-A3B18tok/sQ8_0won't run
GLM 4.7 Flash10tok/sQ8_0won't run
Devstral Small 2 24B4tok/sQ8_0won't run
Nemotron 3.5 Lightning 30B-A3B22tok/sUD-Q6_K_XLwon't run
Kimi Linear 48B-A3B23tok/sQ8_0won't run
gpt-oss 20B21tok/sMXFP412tok/sMXFP4
Qwen3 Coder 30B-A3B9tok/sQ8_0won't run
Qwen3 32B3tok/sQ8_0won't run
Gemma 4 12B10tok/sQ8_06tok/sQ8_0
Qwen 3.5 9B12tok/sQ8_07tok/sQ8_0
Gemma 3 27B4tok/sQ8_0won't run
Mistral Small 3.2 24B4tok/sQ8_0won't run
DeepSeek R1 Distill Qwen 32B3tok/sQ8_0won't run
Llama 3.3 70B2tok/sQ8_0won't run
Ministral 3 14B6tok/sQ8_04tok/sQ5_K_M
Qwen3 14B5tok/sQ8_04tok/sQ5_K_M
DeepSeek R1 0528 Qwen3 8B8tok/sQ8_04tok/sQ8_0
Qwen3 8B8tok/sQ8_04tok/sQ8_0
Phi-4 14B5tok/sQ8_0won't run
Qwen 3.5 4B21tok/sQ8_012tok/sQ8_0
Ministral 3 8B8tok/sQ8_04tok/sQ8_0
Granite 4.2 8B7tok/sQ8_04tok/sQ8_0
Gemma 4 E4B14tok/sQ8_08tok/sQ8_0
LFM2.5 8B-A1B20tok/sQ8_011tok/sQ8_0
Llama 3.1 8B9tok/sQ6_K5tok/sQ6_K
Qwen2.5 0.5B105tok/sQ8_058tok/sQ8_0
Qwen3.5 0.8B86tok/sQ8_048tok/sQ8_0
Qwen3.5 2B62tok/sQ5_K_M35tok/sQ5_K_M
DeepSeek R1 Distill Qwen 1.5B37tok/sQ8_021tok/sQ8_0
MiniCPM5 2B26tok/sQ8_014tok/sQ8_0
Spark X2.5 4B26tok/smlx-4bit14tok/smlx-4bit
Mellum2.1 12B A2.5B Thinking23tok/sQ8_013tok/sQ8_0
FrogNano 4B 260921tok/sQ8_011tok/sQ8_0
LFM2.5 2.6B20tok/sQ8_011tok/sQ8_0
LFM2.5 VL 3B20tok/sQ8_011tok/sQ8_0
Kimi VL A3B Thinking 250616tok/sQ8_09tok/sQ8_0
Kimi VL A3B16tok/smlx-8bit9tok/smlx-8bit
granite vision 4.1 4b16tok/sQ8_09tok/sQ8_0
granite 4.2 3b15tok/sQ8_08tok/sQ8_0
GLM 4 9B 041413tok/smlx-5bit7tok/smlx-5bit
DeepSeek R1 Distill Qwen 7B12tok/sQ8_07tok/sQ8_0
LFM2.5 24B A2B11tok/sQ8_07tok/sQ4_K_M
GLM 4.6V Flash11tok/sQ8_06tok/sQ8_0
GLM 4.1V 9B Thinking10tok/smlx-8bit5tok/smlx-8bit
DeepSeek R1 Distill Llama 8B8tok/sQ8_05tok/sQ8_0
Ministral 3 8B Reasoning 25128tok/sQ8_04tok/sQ8_0
Qwen3.8 27B OBLITERATED6tok/smlx-4bitwon't run
DeepSeek R1 Distill Qwen 14B5tok/sQ8_0won't run
GLM 4.5V4tok/sQ4_K_Mwon't run
Devstral Small 25074tok/sQ8_0won't run
GLM 4.6V4tok/sQ4_K_Mwon't run
Magistral Small 25064tok/sQ8_0won't run
Magistral Small 25074tok/sQ8_0won't run
Magistral Small 25094tok/sQ8_0won't run
granite 4.2 30b3tok/sQ8_0won't run
DeepSeek R1 Distill Llama 70B2tok/sQ8_0won't run
Devstral 2 123B Instruct 25121tok/sQ4_K_Mwon't run
Kimi Dev 72B1tok/sQ8_0won't run
Mistral Medium 3.5 128B1tok/sQ4_K_Mwon't run
DeepSeek V4 Flashwon't runwon't run
GLM 5.3 Flashwon't runwon't run
DeepSeek R1won't runwon't run
DeepSeek R1 0528won't runwon't run
DeepSeek R1 Zerowon't runwon't run
DeepSeek V3won't runwon't run
DeepSeek V3 0324won't runwon't run
DeepSeek V3.1won't runwon't run
DeepSeek V3.1 Terminuswon't runwon't run
DeepSeek V3.2won't runwon't run
DeepSeek V4 Flash 0731won't runwon't run
DeepSeek V4 Flash Vision Expwon't runwon't run
GLM 4.5won't runwon't run
GLM 4.6won't runwon't run
GLM 4.7won't runwon't run
GLM 5won't runwon't run
GLM 5.1won't runwon't run
GLM 5.2won't runwon't run
GLM 5.3won't runwon't run
Kimi K2won't runwon't run
Kimi K2 Instruct 0905won't runwon't run
Kimi K2 Thinkingwon't runwon't run
Kimi K2.5won't runwon't run
MiniMax M2won't runwon't run
MiniMax M2.1won't runwon't run
MiniMax M2.5won't runwon't run
MiniMax M3won't runwon't run

Questions

Which is better for local AI, the GMKtec EVO-X2 Ryzen AI Max+ 395 128GB or the Mac mini M4 32GB?
Choose the GMKtec EVO-X2 Ryzen AI Max+ 395 128GB to run bigger models. Choose the GMKtec EVO-X2 Ryzen AI Max+ 395 128GB for speed. Choose the Mac mini M4 32GB to spend less.
Which can run bigger models, the GMKtec EVO-X2 Ryzen AI Max+ 395 128GB or the Mac mini M4 32GB?
At 64k context the GMKtec EVO-X2 Ryzen AI Max+ 395 128GB runs 74 of the models we track fully in memory and the Mac mini M4 32GB 38; only the GMKtec EVO-X2 Ryzen AI Max+ 395 128GB holds Qwen 3.8 Flash Next.
Which is faster for local AI, the GMKtec EVO-X2 Ryzen AI Max+ 395 128GB or the Mac mini M4 32GB?
On the models both run fully in memory, the GMKtec EVO-X2 Ryzen AI Max+ 395 128GB is about 1.8× as fast as the Mac mini M4 32GB (estimated, median across models).
What is the price difference between the GMKtec EVO-X2 Ryzen AI Max+ 395 128GB and the Mac mini M4 32GB?
In the US the GMKtec EVO-X2 Ryzen AI Max+ 395 128GB costs $1,600 at Amazon and the Mac mini M4 32GB $1,420 at eBay: the Mac mini M4 32GB is $180 cheaper.

More comparisons