Apple · M4 · Mini PC
Best AI Models for Mac Mini M4 (32 GB)
Runs 35B MoE models well; 9B, 14B and 27–31B fit but are slow (13, 9 and 5 tok/s), and 70B doesn't fit in 28.5 GB.
| # | Model | Speed | llmrun Score | Quant · Memory |
|---|---|---|---|---|
| 1 | Qwen3.6 35B A3BAlibabaHow to run → Q4_K_M · 22.0 GB · 23.5 t/s | 23.5 tok/s | Q4_K_M · 22.0 GB | |
| 2 | Gemma 4 26B A4B ITGoogle Q4_K_M · 16.5 GB · 25.8 t/s | 25.8 tok/s | Q4_K_M · 16.5 GB | |
| 3 | Qwen3 30B A3B Thinking 2507Alibaba Q4_K_M · 19.4 GB · 29.3 t/s | 29.3 tok/s | Q4_K_M · 19.4 GB | |
| 4 | GPT OSS 20BOpenAI Q4_K_M · 13.1 GB · 28.1 t/s | 28.1 tok/s | Q4_K_M · 13.1 GB | |
| 5 | Q4_K_M · 19.3 GB · 36.2 t/s | 36.2 tok/s | Q4_K_M · 19.3 GB |
| # | Model | Speed | Coding Score | Quant · Memory |
|---|---|---|---|---|
| 1 | Qwen3.6 35B A3BAlibabaHow to run → Q4_K_M · 22.0 GB · 23.5 t/s | 23.5 tok/s | Q4_K_M · 22.0 GB | |
| 2 | Gemma 4 26B A4B ITGoogle Q4_K_M · 16.5 GB · 25.8 t/s | 25.8 tok/s | Q4_K_M · 16.5 GB | |
| 3 | GPT OSS 20BOpenAI Q4_K_M · 13.1 GB · 28.1 t/s | 28.1 tok/s | Q4_K_M · 13.1 GB | |
| 4 | Q4_K_M · 7.9 GB · 15.8 t/s | 15.8 tok/s | Q4_K_M · 7.9 GB | |
| 5 | Q4_K_M · 19.3 GB · 36.2 t/s | 36.2 tok/s | Q4_K_M · 19.3 GB |
| # | Model | Speed | Long context Score | Quant · Memory |
|---|---|---|---|---|
| 1 | Q4_K_M · 16.5 GB · 25.8 t/s | 25.8 tok/s | Q4_K_M · 16.5 GB | |
| 2 | Qwen3 30B A3B Thinking 2507Alibaba Q4_K_M · 19.4 GB · 29.3 t/s | 29.3 tok/s | Q4_K_M · 19.4 GB | |
| 3 | Qwen3.5 4BAlibaba Q4_K_M · 3.4 GB · 25.8 t/s | 25.8 tok/s | Q4_K_M · 3.4 GB | |
| 4 | GPT OSS 20BOpenAI Q4_K_M · 13.1 GB · 28.1 t/s | 28.1 tok/s | Q4_K_M · 13.1 GB | |
| 5 | Q4_K_M · 19.3 GB · 36.2 t/s | 36.2 tok/s | Q4_K_M · 19.3 GB |
| # | Model | Speed ↓ | llmrun Score | Quant · Memory |
|---|---|---|---|---|
| 1 | Q4_K_M · 19.4 GB · 29.3 t/s | 29.3 tok/s | Q4_K_M · 19.4 GB | |
| 2 | Q4_K_M · 19.3 GB · 36.2 t/s | 36.2 tok/s | Q4_K_M · 19.3 GB | |
| 3 | Qwen3 30B A3B Instruct 2507Alibaba Q4_K_M · 19.4 GB · 29.3 t/s | 29.3 tok/s | Q4_K_M · 19.4 GB | |
| 4 | GPT OSS 20BOpenAI Q4_K_M · 13.1 GB · 28.1 t/s | 28.1 tok/s | Q4_K_M · 13.1 GB | |
| 5 | Gemma 4 26B A4B ITGoogle Q4_K_M · 16.5 GB · 25.8 t/s | 25.8 tok/s | Q4_K_M · 16.5 GB |
| # | Model | Speed | Vision Score | Quant · Memory |
|---|---|---|---|---|
| 1 | Qwen3.6 35B A3BAlibabaHow to run → Q4_K_M · 22.0 GB · 23.5 t/s | 23.5 tok/s | Q4_K_M · 22.0 GB | |
| 2 | Gemma 4 26B A4B ITGoogle Q4_K_M · 16.5 GB · 25.8 t/s | 25.8 tok/s | Q4_K_M · 16.5 GB | |
| 3 | Qwen3.5 4BAlibaba Q4_K_M · 3.4 GB · 25.8 t/s | 25.8 tok/s | Q4_K_M · 3.4 GB | |
| 4 | Gemma 3 4B ITGoogle Q4_K_M · 5.0 GB · 29.6 t/s | 29.6 tok/s | Q4_K_M · 5.0 GB | |
| 5 | MiniCPM V 4.7 35B A3BOpenBMB Q4_K_M · 22.1 GB · 26.9 t/s | 26.9 tok/s | Q4_K_M · 22.1 GB |
The Mac Mini M4 is a 32 GB mini PC with more memory than speed: it runs 35B MoE models well. macOS keeps about 3.5 GB, leaving 28.5 GB for models. Prompt reading is its weak spot: a 4K-token prompt waits 23 s before the first word on Qwen3.6 35B A3B. At its $999 MSRP it gives 13.2 tok/s per $1,000, 4th of the 5 machines compared here. M4 generation, released 2024; the newer M5 writes 28% faster with 1.9× the compute.
Mac Mini M4 strengths and weaknesses
Strengths
- Runs 35B MoE models: Qwen3.6 35B A3B at 23.5 tok/s in 22 GB
- Low power draw: 65 W at full load for the whole machine
Weaknesses
- 70B models don't fit: Llama 3.3 70B Instruct needs 51.3 GB at Q4 vs 28.5 GB usable here
- Slow on large dense models: Qwen3.8 27B at 4.8 tok/s
- Slow prompt reading: A 32K-token prompt on Qwen3.6 35B A3B waits 3.0 min for the first token
- No CUDA: vLLM, AWQ, GPTQ and FP8 builds don't run; use GGUF or MLX
Memory for models
28.5 GB of 32 GB70B models need 51.3 GB at Q4, more than the 28.5 GB available.Or rent an A100 (80 GB) from $1.11/hr →
Speed per dollar and per watt
On Qwen3.5 9B (9B params) at Q4_K_M. New means a shop price, either a street quote or the maker's own price; used is compared with used. The last two columns are the speed you get per $1,000 of price and per 100 W of power; higher is better. The Mac Mini M4 has no current new price, so it is shown at its MSRP.
| Machine | New price | tok/s | tok/s per $1,000 | tok/s per 100 W |
|---|---|---|---|---|
Mac Studio M4 Max36 GB | about $2,499 | 45.1 | 18.0 | 31.1 |
Mac Mini M416 GB | about $799 | 13.2 | 16.5 | 20.3 |
| about $2,299 | 33.8 | 14.7 | – | |
Mac Mini M432 GB | about $999 | 13.2 | 13.2 | 20.3 |
MacBook Pro 14 M532 GB | about $1,839 | 16.9 | 9.2 | – |
| Machine | Used price | tok/s | tok/s per $1,000 | tok/s per 100 W |
|---|---|---|---|---|
RTX 409024 GBBest used value | about $3,000 | 103 | 34.3 | 22.9 |
Mac Studio M4 Max36 GB | about $2,000 | 45.1 | 22.6 | 31.1 |
| about $2,109 | 33.8 | 16.0 | – |
Cost per token: the Mac Mini M4 against 16–64 GB machines and GPUs
On Qwen3.5 9B (9B params, Q4_K_M, 4K context). Speed, cost per token and power per token depend on the model.
- Machine, MSRP
- $4.80
- Power
- $0.21
Renting costs as much as buying after ≈1,800 hours (≈1.2 years at 4 h a day).
Each dot is a machine or GPU; hover or tap it for its numbers, click to open it. Machines without a price for the chosen basis are left out. Machines without a published power figure are left out of Per watt and costed without power. How we calculate cost per token
Other memory sizes of the same machine first, then machines and GPUs people weigh against it, each with the one reason it is worth a look. Speeds on Qwen3.5 9B.
Output speed depends on memory bandwidth (120 GB/s). Grey rows need more than the 28.5 GB available.
- 9BQwen3.5 9B · Q4_K_M · 6.6 GB13.2tok/s
- 8.8tok/s
- 4.8tok/s
- 35B MoEQwen3.6 35B A3B · Q4_K_M · 22.0 GB23.5tok/s
- 70BLlama 3.3 70B Instruct · not at 4-bitneeds 51.3 GB
- 120B MoEGPT OSS 120B · not at 4-bitneeds 70.8 GB
Estimated speeds: Q4_K_M, 4K context, one user. How we estimate
Chip generations: writing vs compute
M1 → M5Writing speed follows memory bandwidth and barely moved: 7.5 to 16.9 tok/s since the M1. Compute, which sets how fast prompts are read, grew 3.1×. Indexed to the M1; speeds on Qwen3.5 9B, same 16 GB.
All 1165 models on the Mac Mini M4
Featured onlyBest quant that fits in 28.5 GB, ordered by llmrun Score. Fit grades how well it runs here, S best. How ranking works
| Model | Quant | VRAM | Speed | Context | Score | Fit |
|---|---|---|---|---|---|---|
Qwen3 Next 80B A3B InstructChatFunctions | IQ2_M | 27.9 GB | 31.1 t/s | 262K | C43, Usable | |
| Q4_K_M | 17.8 GB | 5.7 t/s | 41K | C42, Usable | ||
| Q4_K_M | 7.9 GB | 15.9 t/s | — | B56, Decent | ||
Typhoon Ocr1.5 2BVisionFunctions | Q4_K_M | 3.5 GB | 46.4 t/s | 262K | A78, Runs well | |
| Q4_K_M | 2.3 GB | 50.3 t/s | 4K | A80, Runs well | ||
Idefics3 8B Llama3Vision | Q4_K_M | 7.5 GB | 14.9 t/s | 131K | B55, Decent | |
GPT Neo 125MChat | BF16 | 0.3 GB | 254.5 t/s | 2K | S100, Runs great | |
Qwen2.5 0.5B InstructChatFunctions | Q4_K_M | 0.8 GB | 135.5 t/s | 33K | S96, Runs great | |
| Q4_K_M | 3.7 GB | 31.7 t/s | 8K | A71, Runs well | ||
| Q4_K_M | 0.1 GB | 646.2 t/s | 2K | S100, Runs great | ||
Gpt2 LargeChat | Q4_K_M | 0.5 GB | 155.6 t/s | 1K | S97, Runs great | |
H2ovl Mississippi 2BChatVision | BF16 | 6.5 GB | 17.7 t/s | — | B58, Decent | |
Bloom 560MChat | Q4_K_M | 1.6 GB | 227.0 t/s | — | S100, Runs great | |
Eurus 2 7B PRIMEChat | Q4_K_M | 5.1 GB | 16.8 t/s | 4K | B56, Decent | |
Qwen2.5 Omni 3BChat | Q4_K_M | 6.0 GB | 23.0 t/s | — | B63, Decent |
Specs
- Chip
- M4
- CPU cores
- 10
- GPU cores
- 10
- Unified memory
- 32 GB
- For models
- 28.5 GB (macOS keeps 3.5 GB)
- Memory bandwidth
- 120 GB/s
- FP16 (dense)
- 8.6 TFLOPS
- Neural Engine
- 38 TOPS
- Power
- 65 W max
- Released
- 2024-11-08
- MSRP
- $999
Mac Mini M4 FAQ
Is the Mac Mini M4 (32 GB) worth it for local LLMs in 2026?
At its $999 MSRP it gives 13.2 tok/s per $1,000 on Qwen3.5 9B, 4th of the 5 machines compared here. It runs 35B MoE models well; 9B, 14B and 27–31B fit but are slow (13, 9 and 5 tok/s), and 70B doesn't fit in 28.5 GB.
How much memory is available for AI on the Mac Mini M4 (32 GB)?
About 28.5 GB of the 32 GB. macOS keeps roughly 3.5 GB for itself and other apps.
What is the largest model the Mac Mini M4 (32 GB) can run?
Qwen3.6 35B A3B (36.0B) is the largest at 4-bit or better: at Q4_K_M it uses 22.0 GB, 77% of the available memory, and runs at about 23.5 tok/s.
Can the Mac Mini M4 (32 GB) run 70B models?
No. Llama 3.3 70B Instruct needs 51.3 GB at Q4, more than the 28.5 GB available.
How long does the Mac Mini M4 (32 GB) take to read a long prompt?
On Qwen3.6 35B A3B, about 23 s for a 4K-token prompt and 3.0 min for a 32K-token document before the first word appears.
Is renting a GPU cheaper than buying the Mac Mini M4?
An A100 (40 GB) rents from $0.55/hr. Renting costs as much as buying after ≈1,800 hours (≈1.2 years at 4 h a day).