readygpu

Mac mini M4 32GB vs NVIDIA GeForce RTX 3090

For running models locally, priced in Germany. At 8k context the first runs 26 models fully in memory, the second 28.

In short

The Mac mini M4 32GB has more memory for models: 32 GB against 24 GB. At 8k context the NVIDIA GeForce RTX 3090 runs 28 of the models we track fully in memory and the Mac mini M4 32GB 26; only the NVIDIA GeForce RTX 3090 holds Qwen 3.6 35B-A3B. On the models both run fully in memory, the NVIDIA GeForce RTX 3090 is about 7.4× as fast as the Mac mini M4 32GB (estimated, median across models). In Germany the Mac mini M4 32GB costs 1.400 € at eBay (used) and the NVIDIA GeForce RTX 3090 1.850 € at eBay: the Mac mini M4 32GB is 450 € cheaper.

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

Mac mini M4 32GB

1.400 €

eBay, used (asking)

Add to build
NVIDIA GeForce RTX 3090

1.850 €

eBay

Add to build

Specifications

SpecMac mini M4 32GBNVIDIA GeForce RTX 3090
ChipApple M4 (10-core GPU)–
Memory bus128-bit384-bit
Memory bandwidth120 GB/s936 GB/s
FP16 compute3.76 TFLOPS71 TFLOPS
SoftwareMetalCUDA
Unified memory32 GB–
GPU can use by default21.3 GB–
GPU can use, tuned24 GB–
Operating systemmacos–
ProcessorM4 10-core–
Storage512 GB–
GPU–GeForce RTX 3090
Memory–24 GB GDDR6X
Board power–350 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 mini M4 32GBNVIDIA GeForce RTX 3090
Qwen 3.8 27B4tok/sUD-Q5_K_M26tok/sUD-Q6_K
Qwen 3.6 27B4tok/sQ5_K_M25tok/sQ6_K
Gemma 4 31B4tok/sQ4_K_M25tok/sQ5_K_M
Qwen 3.5 27B4tok/sQ5_K_M25tok/sQ6_K
Qwen 3.6 35B-A3Bwon't run168tok/sUD-Q4_K_M
Gemma 4 26B-A4B13tok/sUD-Q5_K_M88tok/sUD-Q6_K
GLM 4.7 Flash12tok/sQ5_K_M87tok/sQ5_K_M
Devstral Small 2 24B4tok/sQ6_K28tok/sQ6_K
Qwen 3.5 35B-A3Bwon't run168tok/sQ4_K_M
Qwen3 Coder 30B-A3B18tok/sQ4_K_M118tok/sQ5_K_M
gpt-oss 20B18tok/sMXFP4132tok/sMXFP4
Qwen3 32B4tok/sQ4_K_M27tok/sQ4_K_M
Gemma 4 12B6tok/sQ8_044tok/sQ8_0
Qwen 3.5 9B8tok/sQ8_059tok/sQ8_0
Gemma 3 27B4tok/sQ5_K_M25tok/sQ6_K
Mistral Small 3.2 24B4tok/sQ6_K28tok/sQ6_K
DeepSeek R1 Distill Qwen 32B4tok/sQ4_K_M26tok/sQ4_K_M
Ministral 3 14B5tok/sQ8_037tok/sQ8_0
Qwen3 14B5tok/sQ8_034tok/sQ8_0
DeepSeek R1 0528 Qwen3 8B8tok/sQ8_059tok/sQ8_0
Qwen3 8B8tok/sQ8_059tok/sQ8_0
Phi-4 14B5tok/sQ8_034tok/sQ8_0
Qwen 3.5 4B16tok/sQ8_0121tok/sQ8_0
Ministral 3 8B8tok/sQ8_057tok/sQ8_0
Granite 4.2 8B7tok/sQ8_054tok/sQ8_0
Gemma 4 E4B9tok/sQ8_069tok/sQ8_0
LFM2.5 8B-A1B32tok/sQ8_0237tok/sQ8_0
Llama 3.1 8B10tok/sQ6_K76tok/sQ6_K
Qwen 3.5 122B-A10Bwon't runoffload
Qwen3 Coder Nextwon't runoffload
Mistral Small 4 119Bwon't runoffload
GLM 4.5 Airwon't runoffload
Qwen3 Next 80B-A3Bwon't runoffload
Nemotron 3.5 Lightning 30B-A3Bwon't runoffload
Kimi Linear 48B-A3Bwon't runoffload
Llama 3.3 70Bwon't runoffload
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 Mac mini M4 32GB or the NVIDIA GeForce RTX 3090?
Choose the NVIDIA GeForce RTX 3090 to run bigger models. Choose the NVIDIA GeForce RTX 3090 for speed. Choose the Mac mini M4 32GB to spend less.
Which can run bigger models, the Mac mini M4 32GB or the NVIDIA GeForce RTX 3090?
At 8k context the NVIDIA GeForce RTX 3090 runs 28 of the models we track fully in memory and the Mac mini M4 32GB 26; only the NVIDIA GeForce RTX 3090 holds Qwen 3.6 35B-A3B.
Which is faster for local AI, the Mac mini M4 32GB or the NVIDIA GeForce RTX 3090?
On the models both run fully in memory, the NVIDIA GeForce RTX 3090 is about 7.4× as fast as the Mac mini M4 32GB (estimated, median across models).
What is the price difference between the Mac mini M4 32GB and the NVIDIA GeForce RTX 3090?
In Germany the Mac mini M4 32GB costs 1.400 € at eBay (used) and the NVIDIA GeForce RTX 3090 1.850 € at eBay: the Mac mini M4 32GB is 450 € cheaper.

More comparisons