readygpu

Framework Desktop Ryzen AI Max+ 395 64GB vs Mac Studio M4 Max 64GB

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

In short

Both have 64 GB of memory for models; the Framework Desktop Ryzen AI Max+ 395 64GB has 256 GB/s of bandwidth and the Mac Studio M4 Max 64GB 546 GB/s. At 8k context both run the same 36 models we track fully in memory. On the models both run fully in memory, the Mac Studio M4 Max 64GB is about 2.5× 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 Studio M4 Max 64GB

4.183 €

eBay

Add to build

Specifications

SpecFramework Desktop Ryzen AI Max+ 395 64GBMac Studio M4 Max 64GB
ChipRyzen AI Max+ 395 (Radeon 8060S)Apple M4 Max (40-core GPU)
Memory bus256-bit512-bit
Memory bandwidth256 GB/s546 GB/s
FP16 compute59.4 TFLOPS16.16 TFLOPS
SoftwareROCmMetal
Unified memory64 GB64 GB
GPU can use by default48 GB48 GB
GPU can use, tuned56 GB56 GB
Operating systemlinuxmacos
ProcessorRyzen AI Max+ 395M4 Max 16-core
Storage1000 GB1000 GB

What each runs

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

ModelFramework Desktop Ryzen AI Max+ 395 64GBMac Studio M4 Max 64GB
Qwen 3.8 27B5tok/sQ8_012tok/sQ8_0
Qwen 3.5 122B-A10B23tok/sUD-Q2_K_XL58tok/sUD-Q2_K_XL
Qwen 3.6 27B5tok/sQ8_012tok/sQ8_0
Qwen3 Coder Next32tok/sQ4_K_M81tok/sQ4_K_M
Gemma 4 31B4tok/sQ8_010tok/sQ8_0
Mistral Small 4 119B33tok/sUD-Q2_K_XL83tok/sUD-Q2_K_XL
Qwen 3.6 35B-A3B26tok/sQ8_064tok/sQ8_0
Qwen 3.5 27B5tok/sQ8_012tok/sQ8_0
Gemma 4 26B-A4B19tok/sQ8_047tok/sQ8_0
GLM 4.5 Air10tok/sQ2_K25tok/sQ2_K
Qwen3 Next 80B-A3B41tok/sQ4_K_M103tok/sQ4_K_M
Qwen 3.5 35B-A3B26tok/sQ8_064tok/sQ8_0
GLM 4.7 Flash15tok/sQ8_037tok/sQ8_0
Devstral Small 2 24B5tok/sQ8_013tok/sQ8_0
Nemotron 3.5 Lightning 30B-A3B25tok/sUD-Q6_K_XL62tok/sUD-Q6_K_XL
Kimi Linear 48B-A3B33tok/sQ6_K83tok/sQ6_K
gpt-oss 20B32tok/sMXFP481tok/sMXFP4
Qwen3 Coder 30B-A3B21tok/sQ8_053tok/sQ8_0
Qwen3 32B4tok/sQ8_010tok/sQ8_0
Gemma 4 12B11tok/sQ8_027tok/sQ8_0
Qwen 3.5 9B14tok/sQ8_036tok/sQ8_0
Mistral Small 3.2 24B5tok/sQ8_013tok/sQ8_0
Gemma 3 27B5tok/sQ8_012tok/sQ8_0
DeepSeek R1 Distill Qwen 32B4tok/sQ8_010tok/sQ8_0
Llama 3.3 70B3tok/sQ4_K_M8tok/sQ4_K_M
Ministral 3 14B9tok/sQ8_023tok/sQ8_0
Qwen3 14B8tok/sQ8_021tok/sQ8_0
DeepSeek R1 0528 Qwen3 8B14tok/sQ8_036tok/sQ8_0
Qwen3 8B14tok/sQ8_036tok/sQ8_0
Phi-4 14B8tok/sQ8_021tok/sQ8_0
Qwen 3.5 4B29tok/sQ8_074tok/sQ8_0
Ministral 3 8B14tok/sQ8_035tok/sQ8_0
Granite 4.2 8B13tok/sQ8_033tok/sQ8_0
Gemma 4 E4B17tok/sQ8_042tok/sQ8_0
LFM2.5 8B-A1B58tok/sQ8_0145tok/sQ8_0
Llama 3.1 8B18tok/sQ6_K46tok/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 Studio M4 Max 64GB?
Choose the Mac Studio M4 Max 64GB for speed.
Which can run bigger models, the Framework Desktop Ryzen AI Max+ 395 64GB or the Mac Studio M4 Max 64GB?
At 8k context both run the same 36 models we track fully in memory.
Which is faster for local AI, the Framework Desktop Ryzen AI Max+ 395 64GB or the Mac Studio M4 Max 64GB?
On the models both run fully in memory, the Mac Studio M4 Max 64GB is about 2.5× as fast as the Framework Desktop Ryzen AI Max+ 395 64GB (estimated, median across models).

More comparisons