readygpu

Mac mini M4 16GB vs NVIDIA GeForce RTX 4070

For running models locally, priced in Canada. At 32k context the first runs 27 models fully in memory, the second 28.

In short

The Mac mini M4 16GB has more memory for models: 16 GB against 12 GB. At 32k context the NVIDIA GeForce RTX 4070 runs 28 of the models we track fully in memory and the Mac mini M4 16GB 27; only the NVIDIA GeForce RTX 4070 holds Granite 4.2 8B. On the models both run fully in memory, the NVIDIA GeForce RTX 4070 is about 4.0× as fast as the Mac mini M4 16GB (estimated, median across models). In Canada the Mac mini M4 16GB costs $1,200 at eBay (used) and the NVIDIA GeForce RTX 4070 $1,230 at Amazon: the Mac mini M4 16GB is $30 cheaper.

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

Mac mini M4 16GB

$1,200

eBay, used (asking)

Add to build
NVIDIA GeForce RTX 4070

$1,230

Amazon

Add to build

Specifications

SpecMac mini M4 16GBNVIDIA GeForce RTX 4070
ChipApple M4 (10-core GPU)–
Memory bus128-bit192-bit
Memory bandwidth120 GB/s504.2 GB/s
FP16 compute3.76 TFLOPS29.146 TFLOPS
SoftwareMetalCUDA
Unified memory16 GB–
GPU can use by default10.7 GB–
GPU can use, tuned10.666666666666666 GB–
Operating systemmacos–
ProcessorM4 10-core–
Storage256 GB–
GPU–GeForce RTX 4070
Memory–12 GB GDDR6X
Board power–200 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 16GBNVIDIA GeForce RTX 4070
Gemma 4 12B8tok/sQ5_K_M29tok/sQ6_K
Qwen 3.5 9B9tok/sQ6_K29tok/sQ8_0
DeepSeek R1 0528 Qwen3 8B8tok/sQ4_K_M29tok/sQ5_K_M
Qwen3 8B8tok/sQ4_K_M29tok/sQ5_K_M
Qwen 3.5 4B14tok/sQ8_056tok/sQ8_0
Ministral 3 8B7tok/sQ5_K_M29tok/sQ5_K_M
Granite 4.2 8Bwon't run29tok/sQ4_K_M
Gemma 4 E4B9tok/sQ8_034tok/sQ8_0
LFM2.5 8B-A1B18tok/sQ8_070tok/sQ8_0
Llama 3.1 8B8tok/sQ5_K_M29tok/sQ6_K
Qwen2.5 0.5B84tok/sQ8_0335tok/sQ8_0
Qwen3.5 0.8B63tok/sQ8_0254tok/sQ8_0
Qwen3.5 2B42tok/sQ5_K_M168tok/sQ5_K_M
DeepSeek R1 Distill Qwen 1.5B28tok/sQ8_0110tok/sQ8_0
MiniCPM5 2B19tok/sQ8_077tok/sQ8_0
Mellum2.1 12B A2.5B Thinking18tok/sQ5_K_M73tok/sQ5_K_M
Spark X2.5 4B18tok/smlx-4bit73tok/smlx-4bit
LFM2.5 2.6B16tok/sQ8_064tok/sQ8_0
LFM2.5 VL 3B16tok/sQ8_064tok/sQ8_0
FrogNano 4B 260914tok/sQ8_054tok/sQ8_0
granite vision 4.1 4b12tok/sQ8_050tok/sQ8_0
granite 4.2 3b12tok/sQ8_048tok/sQ8_0
Ministral 3 3B Instruct 251211tok/sQ8_044tok/sQ8_0
GLM 4.6V Flash10tok/sQ4_K_M42tok/sQ4_K_M
GLM 4 9B 04148tok/smlx-5bit33tok/smlx-5bit
DeepSeek R1 Distill Llama 8B8tok/sQ5_K_M29tok/sQ6_K
DeepSeek R1 Distill Qwen 7B8tok/sQ8_031tok/sQ8_0
Ministral 3 8B Reasoning 25127tok/sQ5_K_M29tok/sQ5_K_M
Qwen 3.8 27Bwon't runoffload
Qwen 3.6 27Bwon't runoffload
Gemma 4 31Bwon't runoffload
Qwen 3.5 27Bwon't runoffload
Qwen 3.6 35B-A3Bwon't runoffload
Gemma 4 26B-A4Bwon't runoffload
Devstral Small 2 24Bwon't runoffload
GLM 4.7 Flashwon't runoffload
Qwen 3.5 35B-A3Bwon't runoffload
Nemotron 3.5 Lightning 30B-A3Bwon't runoffload
Kimi Linear 48B-A3Bwon't runoffload
Qwen3 Coder 30B-A3Bwon't runoffload
gpt-oss 20Bwon't runoffload
Qwen3 32Bwon't runoffload
DeepSeek R1 Distill Qwen 32Bwon't runoffload
Gemma 3 27Bwon't runoffload
Mistral Small 3.2 24Bwon't runoffload
Ministral 3 14Bwon't runoffload
Qwen3 14Bwon't runoffload
Phi-4 14Bwon't runoffload
DeepSeek R1 Distill Qwen 14Bwon't runoffload
Devstral Small 2507won't runoffload
GLM 4.1V 9B Thinkingwon't runoffload
Kimi VL A3Bwon't runoffload
Kimi VL A3B Thinking 2506won't runoffload
LFM2.5 24B A2Bwon't runoffload
Magistral Small 2506won't runoffload
Magistral Small 2507won't runoffload
Magistral Small 2509won't runoffload
Ministral 3 14B Reasoning 2512won't runoffload
Muse Glimmer 30Bwon't runoffload
Qwen3.8 27B OBLITERATEDwon't runoffload
granite 4.2 30bwon'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
Qwen 3.5 122B-A10Bwon't runwon't run
Nemotron 3 Super 120B-A12Bwon't runwon't run
Qwen3 Coder Nextwon't runwon't run
gpt-oss 120Bwon't runwon't run
Mistral Small 4 119Bwon't runwon't run
GLM 4.5 Airwon't runwon't run
Qwen3 Next 80B-A3Bwon'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 mini M4 16GB or the NVIDIA GeForce RTX 4070?
Choose the NVIDIA GeForce RTX 4070 to run bigger models. Choose the NVIDIA GeForce RTX 4070 for speed. Choose the Mac mini M4 16GB to spend less.
Which can run bigger models, the Mac mini M4 16GB or the NVIDIA GeForce RTX 4070?
At 32k context the NVIDIA GeForce RTX 4070 runs 28 of the models we track fully in memory and the Mac mini M4 16GB 27; only the NVIDIA GeForce RTX 4070 holds Granite 4.2 8B.
Which is faster for local AI, the Mac mini M4 16GB or the NVIDIA GeForce RTX 4070?
On the models both run fully in memory, the NVIDIA GeForce RTX 4070 is about 4.0× as fast as the Mac mini M4 16GB (estimated, median across models).
What is the price difference between the Mac mini M4 16GB and the NVIDIA GeForce RTX 4070?
In Canada the Mac mini M4 16GB costs $1,200 at eBay (used) and the NVIDIA GeForce RTX 4070 $1,230 at Amazon: the Mac mini M4 16GB is $30 cheaper.

More comparisons

More hardware for local AI