readygpu

Mac mini M4 32GB vs NVIDIA Tesla P40

For running models locally, priced in the US. At 32k context the first runs 49 models fully in memory, the second 53.

In short

The Mac mini M4 32GB has more memory for models: 32 GB against 24 GB. At 32k context the NVIDIA Tesla P40 runs 53 of the models we track fully in memory and the Mac mini M4 32GB 49; only the NVIDIA Tesla P40 holds Qwen 3.6 35B-A3B. On the models both run fully in memory, the NVIDIA Tesla P40 is about 2.7× as fast as the Mac mini M4 32GB (estimated, median across models). In the US the Mac mini M4 32GB costs $1,420 at eBay and the NVIDIA Tesla P40 $345 at Amazon: the NVIDIA Tesla P40 is $1,075 cheaper.

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

Mac mini M4 32GB

$1,420

eBay

Add to build
NVIDIA Tesla P40

$345

Amazon

Add to build

Specifications

SpecMac mini M4 32GBNVIDIA Tesla P40
ChipApple M4 (10-core GPU)–
Memory bus128-bit384-bit
Memory bandwidth120 GB/s346 GB/s
FP16 compute3.76 TFLOPS11.8 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–Tesla P40
Memory–24 GB GDDR5
Board power–250 W
PCIe generation–Gen 3
PCIe lanes–x16

What each runs

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

ModelMac mini M4 32GBNVIDIA Tesla P40
Qwen 3.8 27B4tok/sQ4_K_M9tok/sQ5_K_M
Qwen 3.6 27B4tok/sQ5_K_M10tok/sQ5_K_M
Gemma 4 31B4tok/sQ4_K_M10tok/sQ4_K_M
Qwen 3.5 27B4tok/sQ5_K_M10tok/sQ5_K_M
Qwen 3.6 35B-A3Bwon't run50tok/sQ4_K_M
Gemma 4 26B-A4B12tok/sQ5_K_M29tok/sUD-Q6_K
GLM 4.7 Flash10tok/sQ4_K_M24tok/sQ5_K_M
Devstral Small 2 24B4tok/sQ4_K_M10tok/sQ5_K_M
Qwen 3.5 35B-A3Bwon't run50tok/sQ4_K_M
Nemotron 3.5 Lightning 30B-A3Bwon't run52tok/sMXFP4
gpt-oss 20B15tok/sMXFP440tok/sMXFP4
Qwen3 Coder 30B-A3B9tok/sQ4_K_M26tok/sQ4_K_M
Gemma 4 12B6tok/sQ8_016tok/sQ8_0
Qwen 3.5 9B7tok/sQ8_020tok/sQ8_0
Gemma 3 27B4tok/sQ4_K_M10tok/sQ5_K_M
Mistral Small 3.2 24B4tok/sQ4_K_M10tok/sQ5_K_M
Ministral 3 14B4tok/sQ8_011tok/sQ8_0
Qwen3 14B4tok/sQ8_010tok/sQ8_0
DeepSeek R1 0528 Qwen3 8B6tok/sQ8_016tok/sQ8_0
Qwen3 8B6tok/sQ8_016tok/sQ8_0
Phi-4 14B4tok/sQ6_K10tok/sQ8_0
Qwen 3.5 4B14tok/sQ8_038tok/sQ8_0
Ministral 3 8B6tok/sQ8_016tok/sQ8_0
Granite 4.2 8B5tok/sQ8_015tok/sQ8_0
Gemma 4 E4B9tok/sQ8_023tok/sQ8_0
LFM2.5 8B-A1B18tok/sQ8_048tok/sQ8_0
Llama 3.1 8B7tok/sQ6_K20tok/sQ6_K
Qwen2.5 0.5B84tok/sQ8_0230tok/sQ8_0
Qwen3.5 0.8B63tok/sQ8_0174tok/sQ8_0
Qwen3.5 2B42tok/sQ5_K_M116tok/sQ5_K_M
DeepSeek R1 Distill Qwen 1.5B28tok/sQ8_076tok/sQ8_0
MiniCPM5 2B19tok/sQ8_053tok/sQ8_0
Spark X2.5 4B18tok/smlx-4bit50tok/smlx-4bit
LFM2.5 2.6B16tok/sQ8_044tok/sQ8_0
LFM2.5 VL 3B16tok/sQ8_044tok/sQ8_0
Mellum2.1 12B A2.5B Thinking15tok/sQ8_041tok/sQ8_0
granite vision 4.1 4b12tok/sQ8_034tok/sQ8_0
granite 4.2 3b12tok/sQ8_033tok/sQ8_0
Kimi VL A3B Thinking 250611tok/sQ8_031tok/sQ8_0
LFM2.5 24B A2B11tok/sQ5_K_M29tok/sQ6_K
Kimi VL A3B11tok/smlx-8bit30tok/smlx-8bit
GLM 4 9B 04148tok/smlx-5bit22tok/smlx-5bit
DeepSeek R1 Distill Qwen 7B8tok/sQ8_022tok/sQ8_0
GLM 4.6V Flash7tok/sQ8_019tok/sQ8_0
DeepSeek R1 Distill Llama 8B6tok/sQ8_017tok/sQ8_0
GLM 4.1V 9B Thinking6tok/smlx-8bit16tok/smlx-8bit
Ministral 3 8B Reasoning 25126tok/sQ8_016tok/sQ8_0
DeepSeek R1 Distill Qwen 14B4tok/sQ6_K10tok/sQ8_0
Devstral Small 25074tok/sQ4_K_M10tok/sQ5_K_M
Magistral Small 25064tok/sQ4_K_M10tok/sQ5_K_M
Magistral Small 25074tok/sQ4_K_M10tok/sQ5_K_M
Magistral Small 25094tok/sQ4_K_M10tok/sQ5_K_M
Qwen3.8 27B OBLITERATEDwon't run9tok/smlx-4bit
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
Kimi Linear 48B-A3Bwon't runoffload
Qwen3 32Bwon't runoffload
DeepSeek R1 Distill Qwen 32Bwon'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
Nemotron 3 Super 120B-A12Bwon't runwon't run
gpt-oss 120Bwon'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 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 4.5Vwon'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 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 32GB or the NVIDIA Tesla P40?
Choose the NVIDIA Tesla P40 to run bigger models. Choose the NVIDIA Tesla P40 for speed. Choose the NVIDIA Tesla P40 to spend less.
Which can run bigger models, the Mac mini M4 32GB or the NVIDIA Tesla P40?
At 32k context the NVIDIA Tesla P40 runs 53 of the models we track fully in memory and the Mac mini M4 32GB 49; only the NVIDIA Tesla P40 holds Qwen 3.6 35B-A3B.
Which is faster for local AI, the Mac mini M4 32GB or the NVIDIA Tesla P40?
On the models both run fully in memory, the NVIDIA Tesla P40 is about 2.7× 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 Tesla P40?
In the US the Mac mini M4 32GB costs $1,420 at eBay and the NVIDIA Tesla P40 $345 at Amazon: the NVIDIA Tesla P40 is $1,075 cheaper.

More comparisons