readygpu

ASUS Ascent GX10 128GB vs Mac Studio M3 Ultra 256GB

For running models locally, priced in Germany. At 128k context the first runs 76 models fully in memory, the second 81.

In short

The Mac Studio M3 Ultra 256GB has more memory for models: 256 GB against 128 GB. At 128k context the Mac Studio M3 Ultra 256GB runs 81 of the models we track fully in memory and the ASUS Ascent GX10 128GB 76; only the Mac Studio M3 Ultra 256GB holds DeepSeek V4 Flash 0731. On the models both run fully in memory, the Mac Studio M3 Ultra 256GB is about 3.1× as fast as the ASUS Ascent GX10 128GB (estimated, median across models). In Germany the ASUS Ascent GX10 128GB costs 10.362 € at Amazon and the Mac Studio M3 Ultra 256GB 15.555 € at eBay: the ASUS Ascent GX10 128GB is 5.193 € cheaper.

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

ASUS Ascent GX10 128GB

10.362 €

Amazon

Add to build
Mac Studio M3 Ultra 256GB

15.555 €

eBay

Add to build

Specifications

SpecASUS Ascent GX10 128GBMac Studio M3 Ultra 256GB
ChipGB10 Grace BlackwellApple M3 Ultra (80-core GPU)
Memory bus256-bit1024-bit
Memory bandwidth273 GB/s819 GB/s
FP16 compute62 TFLOPS28 TFLOPS
SoftwareCUDAMetal
Unified memory128 GB256 GB
GPU can use by default119 GB192 GB
GPU can use, tuned122 GB248 GB
Operating systemdgx-osmacos
Processor20-core Arm (GB10)M3 Ultra 32-core
Storage1000 GB2000 GB

What each runs

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

ModelASUS Ascent GX10 128GBMac Studio M3 Ultra 256GB
DeepSeek V4 Flash6tok/sUD-Q2_K_XL14tok/sMXFP4
GLM 5.3 Flash10tok/sUD-Q2_K_XL18tok/sUD-Q4_K_XL
Qwen 3.8 Flash Next2tok/sQ4_K_M5tok/sQ8_0
Qwen 3.8 27B5tok/sQ8_014tok/sQ8_0
MiniMax M2.73tok/sUD-Q2_K_XL8tok/sUD-Q5_K_M
Qwen 3.5 122B-A10B9tok/sQ6_K23tok/sQ8_0
Qwen 3.6 27B5tok/sQ8_014tok/sQ8_0
Qwen3 Coder Next14tok/sQ8_043tok/sQ8_0
gpt-oss 120B13tok/sMXFP440tok/sMXFP4
Nemotron 3 Super 120B-A12B8tok/sUD-Q6_K23tok/sQ8_0
Gemma 4 31B4tok/sQ8_012tok/sQ8_0
Mistral Small 4 119B11tok/sUD-Q6_K30tok/sQ8_0
Qwen 3.6 35B-A3B17tok/sQ8_055tok/sQ8_0
Qwen 3.5 27B5tok/sQ8_014tok/sQ8_0
Gemma 4 26B-A4B14tok/sQ8_045tok/sQ8_0
GLM 4.5 Air3tok/sQ6_K8tok/sQ8_0
Qwen 3.5 35B-A3B17tok/sQ8_055tok/sQ8_0
Qwen3 Next 80B-A3B16tok/sQ8_050tok/sQ8_0
GLM 4.7 Flash8tok/sQ8_026tok/sQ8_0
Devstral Small 2 24B4tok/sQ8_011tok/sQ8_0
Nemotron 3.5 Lightning 30B-A3B24tok/sUD-Q6_K_XL76tok/sUD-Q6_K_XL
Kimi Linear 48B-A3B24tok/sQ8_075tok/sQ8_0
gpt-oss 20B18tok/sMXFP457tok/sMXFP4
Qwen3 Coder 30B-A3B6tok/sQ8_020tok/sQ8_0
Qwen3 32B2tok/sQ8_08tok/sQ8_0
Gemma 4 12B11tok/sQ8_035tok/sQ8_0
Qwen 3.5 9B12tok/sQ8_038tok/sQ8_0
Gemma 3 27B4tok/sQ8_013tok/sQ8_0
Mistral Small 3.2 24B4tok/sQ8_011tok/sQ8_0
DeepSeek R1 Distill Qwen 32B2tok/sQ8_08tok/sQ8_0
Llama 3.3 70B1tok/sQ8_05tok/sQ8_0
Ministral 3 14B5tok/sQ8_015tok/sQ8_0
Qwen3 14B5tok/sQ8_014tok/sQ8_0
DeepSeek R1 0528 Qwen3 8B6tok/sQ8_019tok/sQ8_0
Qwen3 8B6tok/sQ8_019tok/sQ8_0
Phi-4 14B4tok/sQ8_013tok/sQ8_0
Qwen 3.5 4B19tok/sQ8_060tok/sQ8_0
Ministral 3 8B6tok/sQ8_020tok/sQ8_0
Granite 4.2 8B6tok/sQ8_017tok/sQ8_0
Gemma 4 E4B14tok/sQ8_044tok/sQ8_0
LFM2.5 8B-A1B14tok/sQ8_043tok/sQ8_0
Llama 3.1 8B7tok/sQ6_K22tok/sQ6_K
Qwen2.5 0.5B79tok/sQ8_0249tok/sQ8_0
Qwen3.5 0.8B69tok/sQ8_0218tok/sQ8_0
Qwen3.5 2B55tok/sQ5_K_M174tok/sQ5_K_M
DeepSeek R1 Distill Qwen 1.5B30tok/sQ8_094tok/sQ8_0
Mellum2.1 12B A2.5B Thinking22tok/sQ8_070tok/sQ8_0
Spark X2.5 4B21tok/smlx-4bit67tok/smlx-4bit
MiniCPM5 2B20tok/sQ8_064tok/sQ8_0
FrogNano 4B 260919tok/sQ8_059tok/sQ8_0
LFM2.5 2.6B16tok/sQ8_049tok/sQ8_0
LFM2.5 VL 3B16tok/sQ8_049tok/sQ8_0
Kimi VL A3B Thinking 250614tok/sQ8_045tok/sQ8_0
Kimi VL A3B14tok/smlx-8bit44tok/smlx-8bit
GLM 4 9B 041412tok/smlx-5bit39tok/smlx-5bit
granite vision 4.1 4b12tok/sQ8_037tok/sQ8_0
granite 4.2 3b12tok/sQ8_036tok/sQ8_0
GLM 4.6V Flash11tok/sQ8_035tok/sQ8_0
DeepSeek R1 Distill Qwen 7B11tok/sQ8_034tok/sQ8_0
GLM 4.1V 9B Thinking10tok/smlx-8bit31tok/smlx-8bit
LFM2.5 24B A2B8tok/sQ8_025tok/sQ8_0
DeepSeek R1 Distill Llama 8B7tok/sQ8_021tok/sQ8_0
Ministral 3 8B Reasoning 25126tok/sQ8_020tok/sQ8_0
Qwen3.8 27B OBLITERATED6tok/smlx-4bit18tok/smlx-4bit
DeepSeek R1 Distill Qwen 14B4tok/sQ8_013tok/sQ8_0
Devstral Small 25074tok/sQ8_011tok/sQ8_0
Magistral Small 25064tok/sQ8_011tok/sQ8_0
Magistral Small 25074tok/sQ8_011tok/sQ8_0
Magistral Small 25094tok/sQ8_011tok/sQ8_0
GLM 4.5V3tok/sQ4_K_M8tok/sQ8_0
GLM 4.6V3tok/sQ4_K_M8tok/sQ8_0
granite 4.2 30b3tok/sQ8_08tok/sQ8_0
DeepSeek R1 Distill Llama 70B1tok/sQ8_05tok/sQ8_0
Devstral 2 123B Instruct 25121tok/sQ4_K_M3tok/sQ8_0
Kimi Dev 72B1tok/sQ8_04tok/sQ8_0
Mistral Medium 3.5 128B1tok/sQ4_K_M3tok/sQ8_0
DeepSeek V4 Flash 0731won't run11tok/sMXFP4
DeepSeek V4 Flash Vision Expwon't run11tok/sMXFP4
MiniMax M2won't run8tok/sQ5_K_M
MiniMax M2.1won't run8tok/sQ5_K_M
MiniMax M2.5won't run8tok/sQ5_K_M
DeepSeek R1won't runwon't run
DeepSeek R1 0528won't runwon't run
DeepSeek R1 Zerowon'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
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 M3won't runwon't run

Questions

Which is better for local AI, the ASUS Ascent GX10 128GB or the Mac Studio M3 Ultra 256GB?
Choose the Mac Studio M3 Ultra 256GB to run bigger models. Choose the Mac Studio M3 Ultra 256GB for speed. Choose the ASUS Ascent GX10 128GB to spend less.
Which can run bigger models, the ASUS Ascent GX10 128GB or the Mac Studio M3 Ultra 256GB?
At 128k context the Mac Studio M3 Ultra 256GB runs 81 of the models we track fully in memory and the ASUS Ascent GX10 128GB 76; only the Mac Studio M3 Ultra 256GB holds DeepSeek V4 Flash 0731.
Which is faster for local AI, the ASUS Ascent GX10 128GB or the Mac Studio M3 Ultra 256GB?
On the models both run fully in memory, the Mac Studio M3 Ultra 256GB is about 3.1× as fast as the ASUS Ascent GX10 128GB (estimated, median across models).
What is the price difference between the ASUS Ascent GX10 128GB and the Mac Studio M3 Ultra 256GB?
In Germany the ASUS Ascent GX10 128GB costs 10.362 € at Amazon and the Mac Studio M3 Ultra 256GB 15.555 € at eBay: the ASUS Ascent GX10 128GB is 5.193 € cheaper.

More comparisons