llmrun
AppleM2Mini PC

Best AI Models for Mac Mini M2 (24 GB)

Memory:24 GB UnifiedBandwidth:100.0 GB/sGPU Cores:10 GPU coresCPU Cores:8 CPU coresNeural Engine:15.8 TOPS

24 GB unified − 3.5 GB OS overhead = 20.5 GB available for AI models

How does this device compare with your machine?

37 of 293 featured models run great on Mac Mini M2 (24 GB) · best pick: Qwen3.8 27B at ~4 tok/s

largest that fits: 36.0B (Qwen3.6 35B A3B, IQ4_XS) · 70B needs CPU offload or more memory

What LLMs Can Mac Mini M2 (24 GB) Run?

1068 models · 229 run great · 198 run well

Featured only

Score: model quality, 100 = average model. Fit: how well it runs on this hardware, S best to F won't fit. Ordered by quality, recency and Fit. How ranking works

Showing compatibility for Mac Mini M2 (24 GB)

LLM models compatible with Mac Mini M2 (24 GB) — ranked by quality, recency and Fit
ModelVRAMScoreFit
Gemma 4 E4B IT OBLITERATED
8.0B·Q4_K_M·13.2 t/s·131K ctx
5.6 GBB52, Decent
GPT OSS 20B Heretic Ara v3
21.5B·Q4_K_M·21.3 t/s·131K ctx
13.5 GBB62, Decent
Tiny LlamaForCausalLM 3.2
1M·Q4_K_M·233.3 t/s·131K ctx
0.5 GBS100, Runs great
ErniePEUnleashed
3.4B·Q4_K_M·27.8 t/s·262K ctx
3.7 GBA68, Runs well
Hy MT2 7B
8.0B·Q4_K_M·13.0 t/s·262K ctx
7.3 GBB52, Decent
Qwen2.5 1.5B Instruct
1.5B·Q4_K_M·70.7 t/s
2.6 GBS86, Runs great
TeleOCR
1.4B·Q4_K_M·55.1 t/s·128K ctx
2.1 GBA82, Runs well
Distil Qwen3 0.6B Text2sql
596M·Q4_K_M·89.7 t/s·41K ctx
1.6 GBS89, Runs great
Gemma 3 4B T1 IT
3.9B·Q4_K_M·23.5 t/s·131K ctx
3.4 GBB64, Decent
CyberPal2.0 20B
20.9B·Q4_K_M·23.9 t/s·131K ctx
13.1 GBB64, Decent
Hy MT2 1.8B
2.0B·Q4_K_M·42.2 t/s·262K ctx
2.6 GBA76, Runs well
Ministral 3 3B Reasoning 2512
4.3B·Q4_K_M·23.3 t/s·262K ctx
4.2 GBB64, Decent
Qwen3.6 28B
28.2B·Q4_K_M·26.1 t/s·262K ctx
17.4 GBB60, Decent
Llama 3.1 Nemotron Nano VL 8B V1
8.7B·Q4_K_M·12.2 t/s
8.4 GBB51, Decent
Gemma 2 9B IT
9.2B·Q4_K_M·11.5 t/s·8K ctx
7.3 GBB50, Decent
Qwen3.6 35B A3B DFlash
35B·Q3_K_M·44.5 t/s·262K ctx
17.6 GBA70, Runs well

Mac Mini M2 (24 GB) has 20.5 GB of memory for models. In the ranking below, Qwen3.8 27B leads at Q4_K_M: it needs 17.9 GB (87% of that memory) and generates about 4 tokens per second. 37 of the 293 featured models earn a Fit S here, meaning they load with room for context and run fast. Decode speed is set mostly by memory bandwidth (100 GB/s), so smaller and mixture-of-experts models are noticeably quicker than dense models of the same quality. 70B-class models need about 46 GB at Q4_K_M, more than the 20.5 GB available, so they require offloading layers to system RAM or a larger-memory setup.

Runs Well

  • Ling 3.0 Tiny (7.9B at Q4_K_M, ~55.1 tok/s)
  • MiniCPM5 2B (2.5B at Q4_K_M, ~36.8 tok/s)
  • Qwen3.5 2B (2.3B at Q4_K_M, ~39.5 tok/s)
  • LFM2.5 8B A1B (8.5B at Q4_K_M, ~49.1 tok/s)
  • ZAYA1 VL 8B (9.7B at Q4_K_M, ~69.1 tok/s)

Challenging

  • Qwen3.8 27B (27.8B at Q4_K_M, ~4 tok/s) fits but uses 87% of memory, so keep context short
  • Gemma 4 26B A4B IT (25.8B at Q4_K_M, ~22.1 tok/s) fits but uses 80% of memory, so keep context short
  • Qwen3.6 35B A3B (36.0B at IQ4_XS, ~22.1 tok/s) fits but uses 97% of memory, so keep context short
  • 70B-class models need ~46 GB at Q4_K_M, so they need CPU offload or more memory
  • 120B-class models need ~77 GB at Q4_K_M, so they need CPU offload or more memory

Mac Mini M2 (24 GB) Specifications

Brand
Apple
Chip
M2
Type
Mini PC
Unified Memory
24 GB
Memory Bandwidth
100.0 GB/s
GPU Cores
10
CPU Cores
8
Neural Engine
15.8 TOPS
Release Date
2023-01-24

Get Started

Ollama (Recommended)

Runs Qwen3.8 27B on Q4_K_M; this is our top choice here.

$curl -fsSL https://ollama.com/install.sh | sh
$ollama run hf.co/unsloth/Qwen3.8-27B-GGUF:Q4_K_M

LM Studio

LM Studio

Download LM Studio, search for "Qwen3.8 27B", and run it with one click.

Prompt Processing

Estimated for Qwen3.8 27B, the compute-bound phase that reads your prompt before the first reply token appears.

25.8tok/s prefill

Short chat

19.8 s

512 tok prompt

Long chat

159 s

4,096 tok prompt

Document / codebase

1270 s

32,768 tok prompt

Prefill is compute-bound and a different number from the decode tok/s shown elsewhere on this page — how prompt processing works →

Performance figures are estimates calibrated as of 2026-09-21 — see calibration basis →

What to do next with Mac Mini M2 (24 GB)

Which model to run first, a faster option with the same memory, and what more memory unlocks. Computed from the rankings and specs above.

Frequently Asked Questions

Can Mac Mini M2 (24 GB) run Qwen3.8 27B?

Yes. The Mac Mini M2 (24 GB) with 24 GB unified memory can run Qwen3.8 27B (Q4_K_M, 17.9 GB, 87% of memory). It is also the table's best pick, at ~4 tok/s. 37 of the 293 featured models get a Fit of S (runs great) here. The largest featured model that fits at Q4 or better is Qwen3.6 35B A3B (36.0B at IQ4_XS). 70B-class models need around 46 GB at Q4_K_M, meaning they need CPU offloading or extra memory beyond what's available in 20.5 GB. Check the compatibility table above for the full list with Fit, VRAM usage and estimated speed. Models are checked against the 20.5 GB available to AI (that is, the 24 GB of RAM minus roughly 3.5 GB for macOS), while Apple Silicon's unified memory allows the GPU to draw from that same pool without copying data.

How much memory is available for AI on Mac Mini M2 (24 GB)?

The Mac Mini M2 (24 GB) has 24 GB unified memory. After macOS reserves ~3.5 GB for the operating system, approximately 20.5 GB is available for AI models. Model rankings here use that 20.5 GB limit instead of the total 24 GB. Unlike discrete GPUs where VRAM is separate from system RAM, a unified-memory device shares one memory pool between the CPU and GPU — this means no data copying overhead, but you share memory with macOS and open apps.

Is Mac Mini M2 (24 GB) good for AI?

With 24 GB unified memory and 100.0 GB/s bandwidth, the Mac Mini M2 (24 GB) is solid for running local AI models. Models that run well include Ling 3.0 Tiny (7.9B at Q4_K_M, ~55.1 tok/s); MiniCPM5 2B (2.5B at Q4_K_M, ~36.8 tok/s); Qwen3.5 2B (2.3B at Q4_K_M, ~39.5 tok/s). Mid-size models run at good quality, and Metal acceleration keeps interactive chat smooth.

What's the best model for Mac Mini M2 (24 GB)?

For the Mac Mini M2 (24 GB), the best pick right now is Qwen3.8 27B at ~4 tok/s. Also running well: Ling 3.0 Tiny (7.9B at Q4_K_M, ~55.1 tok/s); MiniCPM5 2B (2.5B at Q4_K_M, ~36.8 tok/s); Qwen3.5 2B (2.3B at Q4_K_M, ~39.5 tok/s). The best choice depends on your use case: coding assistants benefit from code-tuned models, general chat from instruction-tuned ones. The list is ranked by quality, recency and Fit.

How fast is Mac Mini M2 (24 GB) for AI inference?

With 100.0 GB/s memory bandwidth, the Mac Mini M2 (24 GB) achieves approximately 16 tok/s on a 7B model at Q4_K_M — that's functional for interactive use. A 14B model runs at ~8 tok/s. In the ranking, the top pick Qwen3.8 27B runs at ~4 tok/s. Apple Silicon achieves high efficiency (~70%) thanks to unified memory — there's no PCIe bottleneck between CPU and GPU.

= (100 GB/s ÷ model GB) × efficiency

Apple Silicon achieves ~70% bandwidth efficiency thanks to and Metal acceleration.

Estimated speed on Mac Mini M2 (24 GB)

~4 tok/s
~22 tok/s
~22 tok/s
~11 tok/s

Real-world results typically within ±20%.

Learn more about tok/s estimation →

Can I run AI offline on Mac Mini M2 (24 GB)?

Yes — once you download a model, it runs entirely on the Mac Mini M2 (24 GB) without internet. Applications like Ollama and LM Studio make it straightforward to download, manage, and run models locally. All your conversations stay private on your device with zero data sent to external servers. This is one of the key advantages of local AI: complete privacy, no API costs, and no rate limits.