readygpu

Mac Studio M4 Max 64GB vs NVIDIA DGX Spark 128GB

For running models locally, priced in the UK. At 32k context the first runs 65 models fully in memory, the second 79.

In short

The NVIDIA DGX Spark 128GB has more memory for models: 128 GB against 64 GB. At 32k context the NVIDIA DGX Spark 128GB runs 79 of the models we track fully in memory and the Mac Studio M4 Max 64GB 65; only the NVIDIA DGX Spark 128GB holds DeepSeek V4 Flash. On the models both run fully in memory, the Mac Studio M4 Max 64GB is about 2.1× as fast as the NVIDIA DGX Spark 128GB (estimated, median across models). In the UK the Mac Studio M4 Max 64GB costs £3,399 at eBay and the NVIDIA DGX Spark 128GB £5,400 at eBay: the Mac Studio M4 Max 64GB is £2,001 cheaper.

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

Mac Studio M4 Max 64GB

£3,399

eBay

Add to build
NVIDIA DGX Spark 128GB

£5,400

eBay

Add to build

Specifications

SpecMac Studio M4 Max 64GBNVIDIA DGX Spark 128GB
ChipApple M4 Max (40-core GPU)GB10 Grace Blackwell
Memory bus512-bit256-bit
Memory bandwidth546 GB/s273 GB/s
FP16 compute16.16 TFLOPS62 TFLOPS
SoftwareMetalCUDA
Unified memory64 GB128 GB
GPU can use by default48 GB119 GB
GPU can use, tuned56 GB122 GB
Operating systemmacosdgx-os
ProcessorM4 Max 16-core20-core Arm (GB10)
Storage1000 GB4000 GB

What each runs

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

ModelMac Studio M4 Max 64GBNVIDIA DGX Spark 128GB
DeepSeek V4 Flashwon't run11tok/sUD-Q2_K_XL
GLM 5.3 Flashwon't run11tok/sUD-Q2_K_XL
Qwen 3.8 Flash Nextwon't run2tok/sQ4_K_M
Qwen 3.8 27B11tok/sQ8_05tok/sQ8_0
MiniMax M2.7won't run8tok/sUD-Q3_K_M
Qwen 3.5 122B-A10B50tok/sUD-Q2_K_XL11tok/sQ6_K
Qwen 3.6 27B12tok/sQ8_06tok/sQ8_0
Qwen3 Coder Next66tok/sQ4_K_M20tok/sQ8_0
Gemma 4 31B10tok/sQ8_05tok/sQ8_0
Nemotron 3 Super 120B-A12Bwon't run9tok/sUD-Q6_K
gpt-oss 120Bwon't run23tok/sMXFP4
Mistral Small 4 119B68tok/sUD-Q2_K_XL15tok/sUD-Q6_K
Qwen 3.6 35B-A3B56tok/sQ8_027tok/sQ8_0
Qwen 3.5 27B12tok/sQ8_06tok/sQ8_0
Gemma 4 26B-A4B42tok/sQ8_020tok/sQ8_0
GLM 4.5 Airwon't run5tok/sQ8_0
Qwen3 Next 80B-A3B80tok/sQ4_K_M25tok/sQ8_0
Qwen 3.5 35B-A3B56tok/sQ8_027tok/sQ8_0
GLM 4.7 Flash30tok/sQ8_014tok/sQ8_0
Devstral Small 2 24B12tok/sQ8_06tok/sQ8_0
Nemotron 3.5 Lightning 30B-A3B60tok/sUD-Q6_K_XL28tok/sUD-Q6_K_XL
Kimi Linear 48B-A3B77tok/sQ6_K29tok/sQ8_0
gpt-oss 20B66tok/sMXFP431tok/sMXFP4
Qwen3 Coder 30B-A3B33tok/sQ8_016tok/sQ8_0
Qwen3 32B8tok/sQ8_04tok/sQ8_0
Gemma 4 12B26tok/sQ8_013tok/sQ8_0
Qwen 3.5 9B33tok/sQ8_016tok/sQ8_0
Mistral Small 3.2 24B12tok/sQ8_06tok/sQ8_0
Gemma 3 27B11tok/sQ8_05tok/sQ8_0
DeepSeek R1 Distill Qwen 32B8tok/sQ8_04tok/sQ8_0
Llama 3.3 70Bwon't run2tok/sQ8_0
Ministral 3 14B18tok/sQ8_09tok/sQ8_0
Qwen3 14B17tok/sQ8_08tok/sQ8_0
DeepSeek R1 0528 Qwen3 8B26tok/sQ8_013tok/sQ8_0
Qwen3 8B26tok/sQ8_013tok/sQ8_0
Phi-4 14B16tok/sQ8_08tok/sQ8_0
Qwen 3.5 4B63tok/sQ8_030tok/sQ8_0
Ministral 3 8B26tok/sQ8_013tok/sQ8_0
Granite 4.2 8B24tok/sQ8_012tok/sQ8_0
Gemma 4 E4B39tok/sQ8_019tok/sQ8_0
LFM2.5 8B-A1B80tok/sQ8_038tok/sQ8_0
Llama 3.1 8B33tok/sQ6_K16tok/sQ6_K
Qwen2.5 0.5B380tok/sQ8_0181tok/sQ8_0
Qwen3.5 0.8B288tok/sQ8_0137tok/sQ8_0
Qwen3.5 2B191tok/sQ5_K_M91tok/sQ5_K_M
DeepSeek R1 Distill Qwen 1.5B125tok/sQ8_060tok/sQ8_0
MiniCPM5 2B87tok/sQ8_041tok/sQ8_0
Spark X2.5 4B82tok/smlx-4bit39tok/smlx-4bit
LFM2.5 2.6B73tok/sQ8_035tok/sQ8_0
LFM2.5 VL 3B73tok/sQ8_035tok/sQ8_0
Mellum2.1 12B A2.5B Thinking67tok/sQ8_032tok/sQ8_0
FrogNano 4B 260962tok/sQ8_030tok/sQ8_0
granite vision 4.1 4b56tok/sQ8_027tok/sQ8_0
granite 4.2 3b54tok/sQ8_026tok/sQ8_0
Kimi VL A3B Thinking 250652tok/sQ8_025tok/sQ8_0
Kimi VL A3B50tok/smlx-8bit24tok/smlx-8bit
Ministral 3 3B Instruct 251250tok/sQ8_024tok/sQ8_0
LFM2.5 24B A2B44tok/sQ8_021tok/sQ8_0
GLM 4 9B 041437tok/smlx-5bit18tok/smlx-5bit
DeepSeek R1 Distill Qwen 7B36tok/sQ8_017tok/sQ8_0
GLM 4.6V Flash31tok/sQ8_015tok/sQ8_0
DeepSeek R1 Distill Llama 8B28tok/sQ8_013tok/sQ8_0
GLM 4.1V 9B Thinking27tok/smlx-8bit13tok/smlx-8bit
Ministral 3 8B Reasoning 251226tok/sQ8_013tok/sQ8_0
Ministral 3 14B Reasoning 251218tok/sQ8_09tok/sQ8_0
DeepSeek R1 Distill Qwen 14B16tok/sQ8_08tok/sQ8_0
Qwen3.8 27B OBLITERATED16tok/smlx-4bit7tok/smlx-4bit
Devstral Small 250712tok/sQ8_06tok/sQ8_0
Magistral Small 250612tok/sQ8_06tok/sQ8_0
Magistral Small 250712tok/sQ8_06tok/sQ8_0
Magistral Small 250912tok/sQ8_06tok/sQ8_0
Muse Glimmer 30B11tok/sQ8_05tok/sQ8_0
granite 4.2 30b9tok/sQ8_04tok/sQ8_0
DeepSeek R1 Distill Llama 70Bwon't run2tok/sQ8_0
Devstral 2 123B Instruct 2512won't run2tok/sQ6_K
GLM 4.5Vwon't run5tok/sQ8_0
GLM 4.6Vwon't run5tok/sQ8_0
Kimi Dev 72Bwon't run2tok/sQ8_0
Mistral Medium 3.5 128Bwon't run2tok/sQ6_K
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 Mac Studio M4 Max 64GB or the NVIDIA DGX Spark 128GB?
Choose the NVIDIA DGX Spark 128GB to run bigger models. Choose the Mac Studio M4 Max 64GB for speed. Choose the Mac Studio M4 Max 64GB to spend less.
Which can run bigger models, the Mac Studio M4 Max 64GB or the NVIDIA DGX Spark 128GB?
At 32k context the NVIDIA DGX Spark 128GB runs 79 of the models we track fully in memory and the Mac Studio M4 Max 64GB 65; only the NVIDIA DGX Spark 128GB holds DeepSeek V4 Flash.
Which is faster for local AI, the Mac Studio M4 Max 64GB or the NVIDIA DGX Spark 128GB?
On the models both run fully in memory, the Mac Studio M4 Max 64GB is about 2.1× as fast as the NVIDIA DGX Spark 128GB (estimated, median across models).
What is the price difference between the Mac Studio M4 Max 64GB and the NVIDIA DGX Spark 128GB?
In the UK the Mac Studio M4 Max 64GB costs £3,399 at eBay and the NVIDIA DGX Spark 128GB £5,400 at eBay: the Mac Studio M4 Max 64GB is £2,001 cheaper.

More comparisons

More hardware for local AI