readygpu

GMKtec EVO-X2 Ryzen AI Max+ 395 128GB vs AMD Radeon PRO W7900

For running models locally, priced in Canada. At 8k context the first runs 78 models fully in memory, the second 68.

In short

The GMKtec EVO-X2 Ryzen AI Max+ 395 128GB has more memory for models: 128 GB against 48 GB. At 8k context the GMKtec EVO-X2 Ryzen AI Max+ 395 128GB runs 78 of the models we track fully in memory and the AMD Radeon PRO W7900 68; only the GMKtec EVO-X2 Ryzen AI Max+ 395 128GB holds DeepSeek V4 Flash. On the models both run fully in memory, the AMD Radeon PRO W7900 is about 3.4× as fast as the GMKtec EVO-X2 Ryzen AI Max+ 395 128GB (estimated, median across models). In Canada the GMKtec EVO-X2 Ryzen AI Max+ 395 128GB costs $5,099 at Amazon and the AMD Radeon PRO W7900 $5,500 at eBay (used): the GMKtec EVO-X2 Ryzen AI Max+ 395 128GB is $401 cheaper.

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

GMKtec EVO-X2 Ryzen AI Max+ 395 128GB

$5,099

Amazon

Add to build
AMD Radeon PRO W7900

$5,500

eBay, used (asking)

Add to build

Specifications

SpecGMKtec EVO-X2 Ryzen AI Max+ 395 128GBAMD Radeon PRO W7900
ChipRyzen AI Max+ 395 (Radeon 8060S)–
Memory bus256-bit384-bit
Memory bandwidth256 GB/s864 GB/s
FP16 compute59.4 TFLOPS122.6 TFLOPS
SoftwareROCmROCm
Unified memory128 GB–
GPU can use by default96 GB–
GPU can use, tuned120 GB–
Operating systemwindows–
ProcessorRyzen AI Max+ 395–
Storage2000 GB–
GPU–Radeon PRO W7900
Memory–48 GB GDDR6
Board power–295 W
PCIe generation–Gen 4
PCIe lanes–x16

What each runs

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

ModelGMKtec EVO-X2 Ryzen AI Max+ 395 128GBAMD Radeon PRO W7900
DeepSeek V4 Flash11tok/sUD-Q2_K_XLwon't run
Qwen 3.8 Flash Next3tok/sUD-Q3_K_XLwon't run
MiniMax M2.715tok/sUD-Q2_K_XLoffload
Qwen 3.8 27B5tok/sQ8_016tok/sQ8_0
Qwen 3.5 122B-A10B10tok/sQ6_K77tok/sUD-Q2_K_XL
Qwen 3.6 27B5tok/sQ8_016tok/sQ8_0
gpt-oss 120B25tok/sMXFP4offload
Qwen3 Coder Next19tok/sQ8_0108tok/sQ4_K_M
Nemotron 3 Super 120B-A12B9tok/sQ5_K_Moffload
Gemma 4 31B4tok/sQ8_014tok/sQ8_0
Mistral Small 4 119B14tok/sUD-Q6_K111tok/sUD-Q2_K_XL
Qwen 3.6 35B-A3B26tok/sQ8_086tok/sQ8_0
Qwen 3.5 27B5tok/sQ8_016tok/sQ8_0
Gemma 4 26B-A4B19tok/sQ8_063tok/sQ8_0
GLM 4.5 Air6tok/sQ6_K39tok/sQ2_K
Qwen 3.5 35B-A3B26tok/sQ8_086tok/sQ8_0
Qwen3 Next 80B-A3B25tok/sQ8_0138tok/sQ4_K_M
GLM 4.7 Flash15tok/sQ8_050tok/sQ8_0
Devstral Small 2 24B5tok/sQ8_018tok/sQ8_0
Nemotron 3.5 Lightning 30B-A3B25tok/sUD-Q6_K_XL83tok/sUD-Q6_K_XL
gpt-oss 20B32tok/sMXFP4108tok/sMXFP4
Kimi Linear 48B-A3B26tok/sQ8_0111tok/sQ6_K
Qwen3 Coder 30B-A3B21tok/sQ8_071tok/sQ8_0
Qwen3 32B4tok/sQ8_013tok/sQ8_0
Gemma 4 12B11tok/sQ8_036tok/sQ8_0
Qwen 3.5 9B14tok/sQ8_048tok/sQ8_0
Mistral Small 3.2 24B5tok/sQ8_018tok/sQ8_0
Gemma 3 27B5tok/sQ8_016tok/sQ8_0
DeepSeek R1 Distill Qwen 32B4tok/sQ8_013tok/sQ8_0
Llama 3.3 70B2tok/sQ8_011tok/sQ4_K_M
Ministral 3 14B9tok/sQ8_030tok/sQ8_0
Qwen3 14B8tok/sQ8_028tok/sQ8_0
DeepSeek R1 0528 Qwen3 8B14tok/sQ8_048tok/sQ8_0
Qwen3 8B14tok/sQ8_048tok/sQ8_0
Phi-4 14B8tok/sQ8_028tok/sQ8_0
Qwen 3.5 4B29tok/sQ8_099tok/sQ8_0
Ministral 3 8B14tok/sQ8_047tok/sQ8_0
Granite 4.2 8B13tok/sQ8_045tok/sQ8_0
Gemma 4 E4B17tok/sQ8_056tok/sQ8_0
LFM2.5 8B-A1B58tok/sQ8_0194tok/sQ8_0
Llama 3.1 8B18tok/sQ6_K62tok/sQ6_K
Qwen2.5 0.5B223tok/sQ8_0752tok/sQ8_0
Qwen3.5 0.8B151tok/sQ8_0510tok/sQ8_0
Qwen3.5 2B91tok/sQ5_K_M306tok/sQ5_K_M
DeepSeek R1 Distill Qwen 1.5B66tok/sQ8_0223tok/sQ8_0
MiniCPM5 2B46tok/sQ8_0157tok/sQ8_0
LFM2.5 2.6B42tok/sQ8_0141tok/sQ8_0
LFM2.5 VL 3B42tok/sQ8_0141tok/sQ8_0
Spark X2.5 4B41tok/smlx-4bit140tok/smlx-4bit
granite vision 4.1 4b33tok/sQ8_0111tok/sQ8_0
Ministral 3 3B Instruct 251231tok/sQ8_0105tok/sQ8_0
granite 4.2 3b31tok/sQ8_0104tok/sQ8_0
LFM2.5 24B A2B30tok/sQ8_0102tok/sQ8_0
Mellum2.1 12B A2.5B Thinking30tok/sQ8_0101tok/sQ8_0
FrogNano 4B 260929tok/sQ8_096tok/sQ8_0
Kimi VL A3B Thinking 250625tok/sQ8_085tok/sQ8_0
Kimi VL A3B24tok/smlx-8bit81tok/smlx-8bit
DeepSeek R1 Distill Qwen 7B16tok/sQ8_056tok/sQ8_0
GLM 4 9B 041416tok/smlx-5bit55tok/smlx-5bit
DeepSeek R1 Distill Llama 8B15tok/sQ8_049tok/sQ8_0
Ministral 3 8B Reasoning 251214tok/sQ8_047tok/sQ8_0
GLM 4.6V Flash14tok/sQ8_046tok/sQ8_0
GLM 4.1V 9B Thinking12tok/smlx-8bit39tok/smlx-8bit
Ministral 3 14B Reasoning 25129tok/sQ8_030tok/sQ8_0
GLM 4.5V9tok/sQ4_K_Moffload
DeepSeek R1 Distill Qwen 14B8tok/sQ8_027tok/sQ8_0
GLM 4.6V8tok/sQ4_K_Moffload
Qwen3.8 27B OBLITERATED7tok/smlx-4bit22tok/smlx-4bit
Devstral Small 25075tok/sQ8_018tok/sQ8_0
Magistral Small 25065tok/sQ8_018tok/sQ8_0
Magistral Small 25075tok/sQ8_018tok/sQ8_0
Magistral Small 25095tok/sQ8_018tok/sQ8_0
Muse Glimmer 30B4tok/sQ8_015tok/sQ8_0
granite 4.2 30b4tok/sQ8_014tok/sQ8_0
DeepSeek R1 Distill Llama 70B2tok/sQ8_011tok/sQ4_K_M
Kimi Dev 72B2tok/sQ8_0offload
Devstral 2 123B Instruct 25122tok/sQ5_K_Mwon't run
Mistral Medium 3.5 128B2tok/sQ5_K_Mwon'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 AMD Radeon PRO W7900?
Choose the GMKtec EVO-X2 Ryzen AI Max+ 395 128GB to run bigger models. Choose the AMD Radeon PRO W7900 for speed. Choose the GMKtec EVO-X2 Ryzen AI Max+ 395 128GB to spend less.
Which can run bigger models, the GMKtec EVO-X2 Ryzen AI Max+ 395 128GB or the AMD Radeon PRO W7900?
At 8k context the GMKtec EVO-X2 Ryzen AI Max+ 395 128GB runs 78 of the models we track fully in memory and the AMD Radeon PRO W7900 68; only the GMKtec EVO-X2 Ryzen AI Max+ 395 128GB holds DeepSeek V4 Flash.
Which is faster for local AI, the GMKtec EVO-X2 Ryzen AI Max+ 395 128GB or the AMD Radeon PRO W7900?
On the models both run fully in memory, the AMD Radeon PRO W7900 is about 3.4× as fast as the GMKtec EVO-X2 Ryzen AI Max+ 395 128GB (estimated, median across models).
What is the price difference between the GMKtec EVO-X2 Ryzen AI Max+ 395 128GB and the AMD Radeon PRO W7900?
In Canada the GMKtec EVO-X2 Ryzen AI Max+ 395 128GB costs $5,099 at Amazon and the AMD Radeon PRO W7900 $5,500 at eBay (used): the GMKtec EVO-X2 Ryzen AI Max+ 395 128GB is $401 cheaper.

More comparisons

More hardware for local AI