





Memory decides which model fits on the machine; the AA score shows how smart that model is; bandwidth decides how fast it answers. The selected configuration has 48GB of memory and 307GB/s of bandwidth.
| Memory | Available to the model | Best model (AA score) | What it can do |
|---|---|---|---|
| 16GB | ~12GB | gpt-oss-20b · no AA v5 score yet | Small models: summarizing, drafting, trials — a handful of users |
| 24GB | ~18GB | Qwen3.8-27B (4-bit, tight on memory) · AA score 20.1 | Small models: summarizing, drafting, trials — a handful of users |
| 32GB | ~24GB | Qwen3.8-27B (4-bit, tight on memory) · AA score 20.1 | Personal use, trying document Q&A |
| 48GB | ~36GB | Qwen3.8-27B · AA score 20.1 | Personal use, trying document Q&A |
| 64GB | ~48GB | Qwen3.8-27B (8-bit) · AA score 20.1 | The 8-bit version, the highest memory tier of Mac mini |
AA score: Artificial Analysis Intelligence Index v5.0, checked 3 Nov 2026 — higher means smarter; the v5 scale is harder, so scores are lower than before. For comparison, Claude Opus 5.5 scores 58.4 and the leader, Gemini 4 Carbon, scores 59.4. Response speed and the number of simultaneous users depend on the GPU and memory bandwidth; Namtech tests on real hardware before advising.
Mac mini suits testing or personal use, not as the AI server of a Namtech internal-AI package.
Up to 24GB · your 48GB configuration
Try out local AI with small models: summarizing, drafting short documents and document Q&A for a few people. A good fit when your business wants to test the waters before a larger investment.
32–48GB · your 48GB configuration
Runs Qwen3.8-27B more comfortably, with longer context: try a document Q&A assistant before rolling one out company-wide.
From 64GB · your 48GB configuration
The highest memory tier of Mac mini: runs Qwen3.8-27B at 8-bit with longer context. If you need an AI server for a whole team or many people asking at once, look at Mac Studio.
There is no fixed user count per tier: it depends on the model, prompt length and how many people ask at once. Namtech measures against your real workload before you commit to a configuration.
| Chip | Apple M5 Pro |
|---|---|
| CPU | 15 cores |
| GPU | 16 cores |
| Neural Engine | 16-core Neural Engine |
| Unified memory | 48GB |
| Memory bandwidth | 307GB/s |
| Storage | 512GB |
| Ports | Back: 3× Thunderbolt 4 (M6, up to 40Gb/s) or 3× Thunderbolt 5 (M5 Pro, up to 120Gb/s) — USB-C, USB 4 up to 40Gb/s, DisplayPort · 1× HDMI (multichannel audio support) · 1× 2.5Gb Ethernet (RJ-45) — configurable to 10Gb Ethernet (Nbase-T) · Front: 2× USB-C (USB 3, up to 10Gb/s) + 3.5 mm headphone jack · Wi-Fi 7 (802.11be), Bluetooth 6, Thread — Apple N1 chip |
| Display support | Up to 3 external displays · M6: two 6K@60Hz displays + one 5K@60Hz display · M5 Pro: three 6K@60Hz displays · with up to two displays: one display up to 8K@60Hz |
| Dimensions | Height 5.0 cm · Width 12.7 cm · Depth 12.7 cm |
| Weight | 0.67 kg or 0.73 kg depending on chip (per Apple’s spec page) |
| Power | 100–240V AC, maximum continuous power 155W · operating temperature 10–35°C |
| In the box | Mac mini and power cord (1.8 m) |
Specifications as published by Apple, updated September 24, 2026. Prices on this page include VAT.
Genuine Apple keyboards, mice, trackpads, cables and Studio Display monitors compatible with Mac Studio and Mac mini. You can order them together with your Mac.
























Original price
VAT included
Details as published by Apple · checked September 24, 2026
| Buy from Namtech | Buy directly from Apple Store | |
|---|---|---|
| Payment | Only a 20% deposit when the order is confirmed; the remaining 80% is paid before delivery | Full payment when ordering, even if you wait weeks for stock |
| Cancellation | Cancel before the order is confirmed: free of charge | Cancel once the order is being prepared for shipping: 20% restocking fee |
| AI advice and setup | Configuration advice based on the AI models you plan to run, help installing MLX, Ollama, LM Studio | Not offered |
| Delivery | Delivered to your door, factory-sealed; you open the seal yourself | Free delivery |
| Invoice | VAT invoice issued by Công ty Cổ phần Trạm Năng Lượng Số | Apple invoice |
| Warranty and returns | Per Apple’s policy | Per Apple’s policy |
The best model the 48GB configuration can run is Qwen3.8-27B, AA score 20.1. Qwen3.8-27B at 4-bit takes about 16.1 GB, Apache 2.0 licence. AA score: Artificial Analysis Intelligence Index v5.0, checked 3 Nov 2026 — higher means smarter; the v5 scale is harder, so scores are lower than before. For comparison, Claude Opus 5.5 scores 58.4 and the leader, Gemini 4 Carbon, scores 59.4. Response speed and the number of simultaneous users depend on the GPU and memory bandwidth; Namtech tests on real hardware before advising.
Mac mini suits testing or personal use, not as the AI server of a Namtech internal-AI package. It is a good machine for trying Qwen3.8-27B before a larger investment, or for a single user. For an AI server for a whole team or many people asking at once, choose Mac Studio (memory up to 512GB; the 512GB version runs GLM-5.3-Flash · AA score 27.8).
Personal use, trying document Q&A. Runs Qwen3.8-27B more comfortably, with longer context: try a document Q&A assistant before rolling one out company-wide. The real number of concurrent users depends on the model and prompt length, so Namtech measures on real hardware against your use case before you commit to a configuration.
71,670,000₫, VAT included, with the default 512GB SSD and 2.5Gb Ethernet. Storage can be upgraded from 512GB to 8TB; 10Gb Ethernet costs an extra 3,060,000₫. The price updates instantly in the configurator at the top of the page.
No. Mac mini memory is built into the chip, so you must choose the right amount when you order. That is why you should work out the models you plan to run before you decide.
Choose 48GB if Qwen3.8-27B · AA score 20.1 is enough for testing or personal use. The Mac mini M5 Pro 15CPU 16GPU 64GB 512GB runs Qwen3.8-27B (8-bit) · AA score 20.1, at 83,900,000₫ (12,230,000₫ more), VAT included. Because RAM cannot be upgraded after purchase, choose the next tier up if you are unsure.
Tell us what you plan to use AI for and how many people will use it — Namtech will recommend a configuration that fits your needs, with no overselling.