A single Mac mini can run the open model Qwen3.8-27B on the machine itself, serving one or two users. Your prompts and documents are processed on your machine and are not sent to an outside AI service. Buy from 40,010,000₫ (VAT included); Namtech advises on configuration and helps with setup.
All three run the best model within Mac mini range. They differ in room left for long context, simultaneous users and speed.
Mostly one user on a tight budget. It runs the model, but more slowly because of lower memory bandwidth.
See specsThe best fit for one or two people: plenty of room for long documents and two people asking at once, with higher memory bandwidth.
See specsEnough memory for the 8-bit version, whose answers are closer to the original than the 4-bit build; it is slower in return.
See specsAA score: Artificial Analysis Intelligence Index v4.3.2, checked 22/10/2026 — higher means smarter; Claude Opus 5.5 scores 57.6 for comparison.
Draft emails, quotes, template contracts and social posts, then edit them yourself.
Summarise contracts, reports and meeting notes; ask questions over your own documents.
Client documents and financial figures are processed on the machine instead of being pasted into an outside AI service.
Explain code, write small functions, fix bugs; connect coding tools through the local API.
| Buy | Rent | |
|---|---|---|
| Where the machine sits | Your office or home | At Namtech; you use it through an API |
| Payment | 20% deposit when the order is confirmed, the remaining 80% before delivery | 12-month term, one month’s deposit, deducted from the first payment |
| Model setup | Namtech helps install MLX, Ollama or LM Studio and a suitable model | Namtech installs, monitors and maintains it |
| Power, network, maintenance | Handled by you | Included in the rental price |
| Invoice | VAT invoice issued by Công ty Cổ phần Trạm Năng Lượng Số | VAT invoice issued by Công ty Cổ phần Trạm Năng Lượng Số |
Warranty and returns for purchased machines follow Apple’s policy. Mac mini rental is temporarily out of stock; you can still leave your details and Namtech will let you know when machines are available.
Mac mini suits testing or personal use, not as the AI server of a Namtech internal-AI package. When more people share it, or you need a larger model, Mac Studio has far more memory and bandwidth.
From 24GB of memory, a Mac mini runs Qwen3.8-27B at 4-bit (about 16.1 GB). The 64GB version runs the 8-bit build (about 29.5 GB). The 16GB version only suits light models such as gpt-oss-20b.
This package is for one or two people. Response speed and the number of simultaneous users depend on the GPU and memory bandwidth; Namtech tests on real hardware before advising. For three people or more, look at Mac Studio.
The model runs on the machine itself; prompts and documents are processed there and not sent to an outside AI service. If you rent, the machine sits at Namtech and you call the model through an API.
Qwen3.8-27B is licensed under Apache 2.0, which allows commercial use.
Buy from 40,010,000₫, or rent from 6,490,000₫/month (12-month term), both including 10% VAT. Mac mini rental is temporarily out of stock.
Leave your details and Namtech will call back with configuration and AI model advice, plus a buy or rent quote.