readygpu

Mac Studio M3 Ultra 96GB vs NVIDIA GeForce RTX 4090

For running models locally, priced in Germany. At 128k context the first runs 69 models fully in memory, the second 32.

In short

The Mac Studio M3 Ultra 96GB has more memory for models: 96 GB against 24 GB. At 128k context the Mac Studio M3 Ultra 96GB runs 69 of the models we track fully in memory and the NVIDIA GeForce RTX 4090 32; only the Mac Studio M3 Ultra 96GB holds Qwen 3.8 27B. On the models both run fully in memory, the NVIDIA GeForce RTX 4090 is about 1.2× as fast as the Mac Studio M3 Ultra 96GB (estimated, median across models). In Germany the Mac Studio M3 Ultra 96GB costs 5.999 € at eBay and the NVIDIA GeForce RTX 4090 6.069 € at Amazon: the Mac Studio M3 Ultra 96GB is 70 € cheaper.

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

Mac Studio M3 Ultra 96GB

5.999 €

eBay

Add to build
NVIDIA GeForce RTX 4090

6.069 €

Amazon

Add to build

Specifications

SpecMac Studio M3 Ultra 96GBNVIDIA GeForce RTX 4090
ChipApple M3 Ultra (80-core GPU)–
Memory bus1024-bit384-bit
Memory bandwidth819 GB/s1,008 GB/s
FP16 compute28 TFLOPS165 TFLOPS
SoftwareMetalCUDA
Unified memory96 GB–
GPU can use by default72 GB–
GPU can use, tuned88 GB–
Operating systemmacos–
ProcessorM3 Ultra 32-core–
Storage1000 GB–
GPU–GeForce RTX 4090
Memory–24 GB GDDR6X
Board power–450 W
PCIe generation–Gen 4
PCIe lanes–x16

What each runs

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

ModelMac Studio M3 Ultra 96GBNVIDIA GeForce RTX 4090
Qwen 3.8 27B14tok/sQ8_0offload
Qwen 3.5 122B-A10B41tok/sQ3_K_Moffload
Qwen 3.6 27B14tok/sQ8_0offload
Qwen3 Coder Next49tok/sQ6_Kwon't run
Nemotron 3 Super 120B-A12B44tok/sUD-Q3_K_Mwon't run
gpt-oss 120B40tok/sMXFP4won't run
Gemma 4 31B12tok/sQ8_0offload
Mistral Small 4 119B44tok/sQ4_K_Moffload
Qwen 3.6 35B-A3B55tok/sQ8_0offload
Qwen 3.5 27B14tok/sQ8_0offload
Gemma 4 26B-A4B45tok/sQ8_064tok/sQ5_K_M
GLM 4.5 Air11tok/sQ2_Kwon't run
Qwen3 Next 80B-A3B56tok/sQ6_Kwon't run
Qwen 3.5 35B-A3B55tok/sQ8_0offload
GLM 4.7 Flash26tok/sQ8_0offload
Devstral Small 2 24B11tok/sQ8_0offload
Nemotron 3.5 Lightning 30B-A3B76tok/sUD-Q6_K_XLoffload
Kimi Linear 48B-A3B75tok/sQ8_0offload
gpt-oss 20B57tok/sMXFP466tok/sMXFP4
Qwen3 Coder 30B-A3B20tok/sQ8_0offload
Qwen3 32B8tok/sQ8_0won't run
Gemma 4 12B35tok/sQ8_041tok/sQ8_0
Qwen 3.5 9B38tok/sQ8_045tok/sQ8_0
Gemma 3 27B13tok/sQ8_0offload
Mistral Small 3.2 24B11tok/sQ8_0offload
DeepSeek R1 Distill Qwen 32B8tok/sQ8_0won't run
Ministral 3 14B15tok/sQ8_0offload
Qwen3 14B14tok/sQ8_0offload
DeepSeek R1 0528 Qwen3 8B19tok/sQ8_0offload
Qwen3 8B19tok/sQ8_0offload
Phi-4 14B13tok/sQ8_0offload
Qwen 3.5 4B60tok/sQ8_071tok/sQ8_0
Ministral 3 8B20tok/sQ8_027tok/sQ4_K_M
Granite 4.2 8B17tok/sQ8_0offload
Gemma 4 E4B44tok/sQ8_052tok/sQ8_0
LFM2.5 8B-A1B43tok/sQ8_050tok/sQ8_0
Llama 3.1 8B22tok/sQ6_K27tok/sQ5_K_M
Qwen2.5 0.5B249tok/sQ8_0292tok/sQ8_0
Qwen3.5 0.8B218tok/sQ8_0256tok/sQ8_0
Qwen3.5 2B174tok/sQ5_K_M204tok/sQ5_K_M
LFM3 1.7B154tok/sQ8_0180tok/sQ8_0
DeepSeek R1 Distill Qwen 1.5B94tok/sQ8_0111tok/sQ8_0
Mellum2.1 12B A2.5B Thinking70tok/sQ8_082tok/sQ8_0
Spark X2.5 4B67tok/smlx-4bit79tok/smlx-4bit
MiniCPM5 2B64tok/sQ8_075tok/sQ8_0
FrogNano 4B 260959tok/sQ8_070tok/sQ8_0
LFM2.5 2.6B49tok/sQ8_057tok/sQ8_0
LFM2.5 VL 3B49tok/sQ8_057tok/sQ8_0
Kimi VL A3B Thinking 250645tok/sQ8_052tok/sQ8_0
Kimi VL A3B44tok/smlx-8bit51tok/smlx-8bit
GLM 4 9B 041439tok/smlx-5bit46tok/smlx-5bit
granite vision 4.1 4b37tok/sQ8_044tok/sQ8_0
granite 4.2 3b36tok/sQ8_043tok/sQ8_0
GLM 4.6V Flash35tok/sQ8_041tok/sQ8_0
DeepSeek R1 Distill Qwen 7B34tok/sQ8_040tok/sQ8_0
GLM 4.1V 9B Thinking31tok/smlx-8bit37tok/smlx-8bit
Ministral 3 3B Instruct 251230tok/sQ8_036tok/sQ8_0
LFM2.5 24B A2B25tok/sQ8_0offload
DeepSeek R1 Distill Llama 8B21tok/sQ8_027tok/sQ5_K_M
Ministral 3 8B Reasoning 251220tok/sQ8_027tok/sQ4_K_M
Qwen3.8 27B OBLITERATED18tok/smlx-4bitoffload
Muse Glimmer 30B16tok/sQ8_029tok/sQ5_K_M
Ministral 3 14B Reasoning 251215tok/sQ8_0offload
DeepSeek R1 Distill Qwen 14B13tok/sQ8_0offload
Devstral Small 250711tok/sQ8_0offload
Magistral Small 250611tok/sQ8_0offload
Magistral Small 250711tok/sQ8_0offload
Magistral Small 250911tok/sQ8_0offload
granite 4.2 30b8tok/sQ8_0won't run
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
Llama 3.3 70Bwon't runwon't run
DeepSeek R1won't runwon't run
DeepSeek R1 0528won't runwon't run
DeepSeek R1 Distill Llama 70Bwon'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
Devstral 2 123B Instruct 2512won't runwon't run
GLM 4.5won't runwon't run
GLM 4.5Vwon't runwon't run
GLM 4.6won't runwon't run
GLM 4.6Vwon'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 Dev 72Bwon'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
Mistral Medium 3.5 128Bwon't runwon't run

Questions

Which is better for local AI, the Mac Studio M3 Ultra 96GB or the NVIDIA GeForce RTX 4090?
Choose the Mac Studio M3 Ultra 96GB to run bigger models. Choose the NVIDIA GeForce RTX 4090 for speed. Choose the Mac Studio M3 Ultra 96GB to spend less.
Which can run bigger models, the Mac Studio M3 Ultra 96GB or the NVIDIA GeForce RTX 4090?
At 128k context the Mac Studio M3 Ultra 96GB runs 69 of the models we track fully in memory and the NVIDIA GeForce RTX 4090 32; only the Mac Studio M3 Ultra 96GB holds Qwen 3.8 27B.
Which is faster for local AI, the Mac Studio M3 Ultra 96GB or the NVIDIA GeForce RTX 4090?
On the models both run fully in memory, the NVIDIA GeForce RTX 4090 is about 1.2× as fast as the Mac Studio M3 Ultra 96GB (estimated, median across models).
What is the price difference between the Mac Studio M3 Ultra 96GB and the NVIDIA GeForce RTX 4090?
In Germany the Mac Studio M3 Ultra 96GB costs 5.999 € at eBay and the NVIDIA GeForce RTX 4090 6.069 € at Amazon: the Mac Studio M3 Ultra 96GB is 70 € cheaper.

More comparisons

More hardware for local AI