readygpu

Framework Desktop Ryzen AI Max+ 395 64GB vs Mac mini M4 Pro 48GB

For running models locally, priced in the UK. At 8k context the first runs 36 models fully in memory, the second 30.

In short

The Framework Desktop Ryzen AI Max+ 395 64GB has more memory for models: 64 GB against 48 GB. At 8k context the Framework Desktop Ryzen AI Max+ 395 64GB runs 36 of the models we track fully in memory and the Mac mini M4 Pro 48GB 30; only the Framework Desktop Ryzen AI Max+ 395 64GB holds Qwen 3.5 122B-A10B. On the models both run fully in memory, the Mac mini M4 Pro 48GB is about 1.3× as fast as the Framework Desktop Ryzen AI Max+ 395 64GB (estimated, median across models).

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

Framework Desktop Ryzen AI Max+ 395 64GB

Not priced here

Add to build
Mac mini M4 Pro 48GB

£3,000

eBay

Add to build

Specifications

SpecFramework Desktop Ryzen AI Max+ 395 64GBMac mini M4 Pro 48GB
ChipRyzen AI Max+ 395 (Radeon 8060S)Apple M4 Pro (20-core GPU)
Memory bus256-bit256-bit
Memory bandwidth256 GB/s273 GB/s
FP16 compute59.4 TFLOPS8.08 TFLOPS
SoftwareROCmMetal
Unified memory64 GB48 GB
GPU can use by default48 GB36 GB
GPU can use, tuned56 GB40 GB
Operating systemlinuxmacos
ProcessorRyzen AI Max+ 395M4 Pro 14-core
Storage1000 GB512 GB

What each runs

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

ModelFramework Desktop Ryzen AI Max+ 395 64GBMac mini M4 Pro 48GB
Qwen 3.8 27B5tok/sQ8_06tok/sQ8_0
Qwen 3.5 122B-A10B23tok/sUD-Q2_K_XLwon't run
Qwen 3.6 27B5tok/sQ8_06tok/sQ8_0
Qwen3 Coder Next32tok/sQ4_K_Mwon't run
Gemma 4 31B4tok/sQ8_05tok/sQ8_0
Mistral Small 4 119B33tok/sUD-Q2_K_XLwon't run
Qwen 3.6 35B-A3B26tok/sQ8_032tok/sQ8_0
Qwen 3.5 27B5tok/sQ8_06tok/sQ8_0
Gemma 4 26B-A4B19tok/sQ8_023tok/sQ8_0
GLM 4.5 Air10tok/sQ2_Kwon't run
Qwen3 Next 80B-A3B41tok/sQ4_K_Mwon't run
Qwen 3.5 35B-A3B26tok/sQ8_032tok/sQ8_0
GLM 4.7 Flash15tok/sQ8_019tok/sQ8_0
Devstral Small 2 24B5tok/sQ8_07tok/sQ8_0
Nemotron 3.5 Lightning 30B-A3B25tok/sUD-Q6_K_XL31tok/sUD-Q6_K_XL
Kimi Linear 48B-A3B33tok/sQ6_K47tok/sQ5_K_M
gpt-oss 20B32tok/sMXFP440tok/sMXFP4
Qwen3 Coder 30B-A3B21tok/sQ8_027tok/sQ8_0
Qwen3 32B4tok/sQ8_05tok/sQ8_0
Gemma 4 12B11tok/sQ8_014tok/sQ8_0
Qwen 3.5 9B14tok/sQ8_018tok/sQ8_0
Mistral Small 3.2 24B5tok/sQ8_07tok/sQ8_0
Gemma 3 27B5tok/sQ8_06tok/sQ8_0
DeepSeek R1 Distill Qwen 32B4tok/sQ8_05tok/sQ8_0
Llama 3.3 70B3tok/sQ4_K_Mwon't run
Ministral 3 14B9tok/sQ8_011tok/sQ8_0
Qwen3 14B8tok/sQ8_010tok/sQ8_0
DeepSeek R1 0528 Qwen3 8B14tok/sQ8_018tok/sQ8_0
Qwen3 8B14tok/sQ8_018tok/sQ8_0
Phi-4 14B8tok/sQ8_010tok/sQ8_0
Qwen 3.5 4B29tok/sQ8_037tok/sQ8_0
Ministral 3 8B14tok/sQ8_017tok/sQ8_0
Granite 4.2 8B13tok/sQ8_017tok/sQ8_0
Gemma 4 E4B17tok/sQ8_021tok/sQ8_0
LFM2.5 8B-A1B58tok/sQ8_073tok/sQ8_0
Llama 3.1 8B18tok/sQ6_K23tok/sQ6_K
DeepSeek V4 Flashwon't runwon't run
GLM 5.3 Flashwon't runwon't run
Qwen 3.8 Flash Nextwon't runwon't run
MiniMax M2.7won't runwon't run
Nemotron 3 Super 120B-A12Bwon't runwon't run
gpt-oss 120Bwon't runwon't run

Questions

Which is better for local AI, the Framework Desktop Ryzen AI Max+ 395 64GB or the Mac mini M4 Pro 48GB?
Choose the Framework Desktop Ryzen AI Max+ 395 64GB to run bigger models. Choose the Mac mini M4 Pro 48GB for speed.
Which can run bigger models, the Framework Desktop Ryzen AI Max+ 395 64GB or the Mac mini M4 Pro 48GB?
At 8k context the Framework Desktop Ryzen AI Max+ 395 64GB runs 36 of the models we track fully in memory and the Mac mini M4 Pro 48GB 30; only the Framework Desktop Ryzen AI Max+ 395 64GB holds Qwen 3.5 122B-A10B.
Which is faster for local AI, the Framework Desktop Ryzen AI Max+ 395 64GB or the Mac mini M4 Pro 48GB?
On the models both run fully in memory, the Mac mini M4 Pro 48GB is about 1.3× as fast as the Framework Desktop Ryzen AI Max+ 395 64GB (estimated, median across models).

More comparisons