readygpu

Mac Studio M3 Ultra 256GB vs NVIDIA DGX Spark 128GB

For running models locally, priced in the US. At 128k context the first runs 42 models fully in memory, the second 42.

In short

The Mac Studio M3 Ultra 256GB has more memory for models: 256 GB against 128 GB. At 128k context both run the same 42 models we track fully in memory. On the models both run fully in memory, the Mac Studio M3 Ultra 256GB is about 3.1× as fast as the NVIDIA DGX Spark 128GB (estimated, median across models). In the US the Mac Studio M3 Ultra 256GB costs $8,499 at eBay and the NVIDIA DGX Spark 128GB $7,399 at eBay: the NVIDIA DGX Spark 128GB is $1,100 cheaper.

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

Mac Studio M3 Ultra 256GB

$8,499

eBay

Add to build
NVIDIA DGX Spark 128GB

$7,399

eBay

Add to build

Specifications

SpecMac Studio M3 Ultra 256GBNVIDIA DGX Spark 128GB
ChipApple M3 Ultra (80-core GPU)GB10 Grace Blackwell
Memory bus1024-bit256-bit
Memory bandwidth819 GB/s273 GB/s
FP16 compute28 TFLOPS62 TFLOPS
SoftwareMetalCUDA
Unified memory256 GB128 GB
GPU can use by default192 GB119 GB
GPU can use, tuned248 GB122 GB
Operating systemmacosdgx-os
ProcessorM3 Ultra 32-core20-core Arm (GB10)
Storage2000 GB4000 GB

What each runs

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

ModelMac Studio M3 Ultra 256GBNVIDIA DGX Spark 128GB
DeepSeek V4 Flash14tok/sUD-Q4_K_XL6tok/sUD-Q2_K_XL
GLM 5.3 Flash18tok/sUD-Q4_K_XL10tok/sUD-Q2_K_XL
Qwen 3.8 Flash Next5tok/sQ8_03tok/sUD-Q4_K_XL
Qwen 3.8 27B14tok/sQ8_05tok/sQ8_0
MiniMax M2.78tok/sUD-Q5_K_M3tok/sUD-Q2_K_XL
Qwen 3.5 122B-A10B23tok/sQ8_09tok/sQ6_K
Qwen 3.6 27B14tok/sQ8_05tok/sQ8_0
Qwen3 Coder Next43tok/sQ8_014tok/sQ8_0
gpt-oss 120B40tok/sMXFP413tok/sMXFP4
Nemotron 3 Super 120B-A12B23tok/sQ8_08tok/sUD-Q6_K
Gemma 4 31B12tok/sQ8_04tok/sQ8_0
Mistral Small 4 119B30tok/sQ8_011tok/sUD-Q6_K
Qwen 3.6 35B-A3B55tok/sQ8_017tok/sQ8_0
Qwen 3.5 27B14tok/sQ8_05tok/sQ8_0
Gemma 4 26B-A4B45tok/sQ8_014tok/sQ8_0
GLM 4.5 Air8tok/sQ8_03tok/sQ6_K
Qwen 3.5 35B-A3B55tok/sQ8_017tok/sQ8_0
Qwen3 Next 80B-A3B50tok/sQ8_016tok/sQ8_0
GLM 4.7 Flash26tok/sQ8_08tok/sQ8_0
Devstral Small 2 24B11tok/sQ8_04tok/sQ8_0
Nemotron 3.5 Lightning 30B-A3B76tok/sUD-Q6_K_XL24tok/sUD-Q6_K_XL
Kimi Linear 48B-A3B75tok/sQ8_024tok/sQ8_0
gpt-oss 20B57tok/sMXFP418tok/sMXFP4
Qwen3 Coder 30B-A3B20tok/sQ8_06tok/sQ8_0
Qwen3 32B8tok/sQ8_02tok/sQ8_0
Gemma 4 12B35tok/sQ8_011tok/sQ8_0
Qwen 3.5 9B38tok/sQ8_012tok/sQ8_0
Gemma 3 27B13tok/sQ8_04tok/sQ8_0
Mistral Small 3.2 24B11tok/sQ8_04tok/sQ8_0
DeepSeek R1 Distill Qwen 32B8tok/sQ8_02tok/sQ8_0
Llama 3.3 70B5tok/sQ8_01tok/sQ8_0
Ministral 3 14B15tok/sQ8_05tok/sQ8_0
Qwen3 14B14tok/sQ8_05tok/sQ8_0
DeepSeek R1 0528 Qwen3 8B19tok/sQ8_06tok/sQ8_0
Qwen3 8B19tok/sQ8_06tok/sQ8_0
Phi-4 14B13tok/sQ8_04tok/sQ8_0
Qwen 3.5 4B60tok/sQ8_019tok/sQ8_0
Ministral 3 8B20tok/sQ8_06tok/sQ8_0
Granite 4.2 8B17tok/sQ8_06tok/sQ8_0
Gemma 4 E4B44tok/sQ8_014tok/sQ8_0
LFM2.5 8B-A1B43tok/sQ8_014tok/sQ8_0
Llama 3.1 8B22tok/sQ6_K7tok/sQ6_K

Questions

Which is better for local AI, the Mac Studio M3 Ultra 256GB or the NVIDIA DGX Spark 128GB?
Choose the Mac Studio M3 Ultra 256GB for speed. Choose the NVIDIA DGX Spark 128GB to spend less.
Which can run bigger models, the Mac Studio M3 Ultra 256GB or the NVIDIA DGX Spark 128GB?
At 128k context both run the same 42 models we track fully in memory.
Which is faster for local AI, the Mac Studio M3 Ultra 256GB or the NVIDIA DGX Spark 128GB?
On the models both run fully in memory, the Mac Studio M3 Ultra 256GB is about 3.1× as fast as the NVIDIA DGX Spark 128GB (estimated, median across models).
What is the price difference between the Mac Studio M3 Ultra 256GB and the NVIDIA DGX Spark 128GB?
In the US the Mac Studio M3 Ultra 256GB costs $8,499 at eBay and the NVIDIA DGX Spark 128GB $7,399 at eBay: the NVIDIA DGX Spark 128GB is $1,100 cheaper.

More comparisons