96GB max RAM · 4 bays · 2 x 10GbE
AINAS Score 81
No verified UK price
View
Can the Minisforum N5 Pro run local AI?
Yes. AMD Ryzen AI 9 HX PRO 370 and a 50 TOPS NPU, backed by a 96GB memory ceiling, lets the Minisforum N5 Pro handle 70B class models at 4-bit with room left for longer context (Estimated).
A graphics card can be added through its OCuLink (external GPU), which is the biggest single speed upgrade for local models.
Local LLM support is documented by the manufacturer.
The N5 Pro is a compact five-bay NAS built around a laptop AI processor. It uses AMD's Ryzen AI 9 HX PRO 370, a 12-core laptop chip with Radeon 890M graphics and a 50 TOPS NPU, in a five-bay chassis that also takes three M.2 drives. Memory is regular DDR5 SO-DIMM with ECC support up to 96GB, so unlike soldered designs you can start small and upgrade.
With the full 96GB fitted it holds 70B class models at 4-bit (Estimated); with 48GB it suits 14B to 32B class models. Minisforum documents running language models on the NPU and GPU, and community testing has shown models running under other operating systems too.
It has two ways to add a GPU: an OCuLink port for an external dock and a PCIe x16 slot wired at x4. Networking is 10GbE plus 5GbE, and there are two USB4 ports. It is sold barebone or with memory bundles and uses an external 280W power brick. It suits home lab owners who want local models, Docker and five bays in a small case, and are comfortable fitting their own memory.
Amazon UK
16GB DDR5 + 128GB SSD
£1,040.41
Verified
In stock
View dealProduct prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon.co.uk at the time of purchase will apply to the purchase of this product.
Each capability shows how it was confirmed: Verified by AINAS, Vendor confirmed, Community confirmed or Unverified.
| Capability | Supported | Evidence | Note |
|---|---|---|---|
| Local LLM | Yes | Vendor confirmed | Minisforum quick start guide describes running an LLM locally on N5 Pro NPU/GPU; LinuxLinks ran Llama 3.2 1B and Qwen3 8B on the NPU via FastFlowLM under Ubuntu |
| Ollama | Yes | Community confirmed | Community Ollama ROCm container tested on N5 Pro under TrueNAS bare metal, not MinisCloud OS |
| Open WebUI | Limited | Unverified | Docker available; no source found showing Open WebUI on N5 Pro |
| RAG | Limited | Unverified | No vendor RAG feature found; possible via Docker |
| AI search | Yes | Vendor confirmed | MinisCloud OS AI photo library with image/semantic search |
| GPU expansion | Yes | Vendor confirmed | OCuLink eGPU plus PCIe x16 slot (PCIe 4.0 x4) |
| Hardware acceleration | Yes | Vendor confirmed | Radeon 890M AV1/H.265 hardware encode/decode, real-time transcoding |
| AI photos | Yes | Vendor confirmed | Face recognition, scene sorting, image search |
| AI video | Limited | Unverified | Page claims NPU enables real-time video analysis; no specific app found |
| Surveillance AI | No | Unverified | No surveillance/NVR app mentioned |
Sold barebone or with memory bundles, the N5 Pro accepts 96GB of DDR5 5600 SO-DIMM including ECC modules. Buy as much as your budget allows: memory decides model size.
Working backwards from 96GB: reserve about 8GB for the operating system and storage services, and at roughly 0.6GB per billion parameters at 4-bit, the largest model that fits is around 146B (Estimated).
ECC support protects against silent memory errors on a box that runs around the clock.
Through OCuLink. An external dock connects a desktop graphics card to the N5 Pro at PCIe speed, which is the most practical upgrade for a compact NAS.
Its 50 TOPS NPU is used by the vendor's own AI features. Open-source LLM runtimes generally ignore NAS NPUs today, so it adds little to chat model speed.
For a personal library, yes. A compact embedding model and a small chat model will run, but the N5 Pro is better as the document store behind a RAG system whose model runs on stronger hardware.
The manufacturer also provides AI search across your files (documented by the manufacturer), which covers simple document questions without setting up your own stack.
Built-in networking: 10GbE + 5GbE. Plenty for a shared knowledge base with several users.
| Configuration | RAM | SSD | Notes |
|---|---|---|---|
| Barebone + 128GB SSD | None | 128GB | Barebone with 128GB system SSD; fit your own DDR5 SO-DIMMs |
| 16GB DDR5 + 128GB SSD | 16GB | 128GB | Global store only; not listed on the UK store |
| 48GB DDR5 ECC + 128GB SSD | 48GB | 128GB | 48GB ECC memory fitted; out of stock on UK and global stores when checked |
| 96GB DDR5 ECC + 128GB SSD | 96GB | 128GB | 96GB ECC memory fitted (the maximum); out of stock when checked |
The closest Minisforum model by AINAS Score. Only the specifications that differ are listed.
Its strongest AINAS profiles are Local LLM at 79 and RAG at 77, which points to people who want a private chat assistant or coding model at home without sending data to a cloud service.
Other Native AI NAS systems closest in price.
Yes. Ollama support is confirmed by community reports. See the capability table above for the source.
Partly. Open WebUI support is not yet verified. See the capability table above for the source.
Yes. AI photos support is documented by the manufacturer. See the capability table above for the source.
No. Surveillance AI support is not yet verified. See the capability table above for the source.
Yes. Hardware acceleration support is documented by the manufacturer. See the capability table above for the source.
It has 10GbE + 5GbE built in, and the 1x PCIe x16 slot (PCIe 4.0 x4 electrical) can take a faster network card. The fastest port is 10GbE.
Every field records its source. Specifications were last checked on . Spot an error? Send a correction with a manufacturer link.