96GB max RAM · 5 bays · 10GbE
AINAS Score 79
£1,040.41
Amazon UK · Verified
Can the Zettlab D8 Ultra run local AI?
It can. The Zettlab D8 Ultra combines Intel Core Ultra 5 125H and a 11 TOPS NPU with as much as 96GB of memory, enough for 70B class models at 4-bit with room left for longer context (Estimated).
Its PCIe x8 slot leaves room to add a GPU when you outgrow CPU inference.
Local LLM support is documented by the manufacturer.
What is the D8 Ultra like?
The Zettlab D8 Ultra is the eight-bay version of the D6 Ultra, with the same Core Ultra 5 125H, the same 32GB of DDR5 expandable to 96GB, two 10GbE ports, two USB4 ports and the external PCIe 4.0 x8 connector. The extra two bays raise maximum raw capacity to around 200TB.
The AI software is the same ZettOS stack: a local assistant with downloadable open models, semantic search across files and media, transcription and photo recognition. We have not found a hands-on review of the D8 Ultra itself, so its AI capability rests on the D6 Ultra reviews and Zettlab's documentation, which is reasonable given the identical platform.
Like the D6 Ultra it is listed as a sold-out pre-order with no confirmed shipping date, and Zettlab does not currently ship to the UK. Choose it over the D6 Ultra only if you need the extra bays; for local AI the two are the same machine. Until it ships, the UGREEN iDX6011 is the nearest eight-bay-class alternative you can actually buy.
Current UK offers
Not currently on Amazon UK
We could not find a UK listing for the Zettlab D8 Ultra from a retailer we track. The specifications below stand on their own; a price appears here as soon as a UK listing is confirmed.
Product 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.
Key specifications
- PCIe
- 1x external SFF-8654 connector (PCIe 4.0 x8); no internal PCIe card slot listed
- Source: Manufacturer
- Docker
- Yes
- VM support
- Yes
- Released
- Not stated
Can the D8 Ultra run local LLMs?
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 | Spec table lists Local LLM and AI; TechRadar saw downloadable Gemma, Phi, QwQ, DeepSeek-R1 and Llama models; XDA says default chat model likely Qwen3-4B |
| Ollama | Limited | Unverified | No Ollama app or ZettOS guide found; Docker with Compose is supported so it is plausible via container |
| Open WebUI | Limited | Unverified | No Open WebUI app or guide found; plausible via Docker |
| RAG | Yes | Vendor confirmed | ZettAI learns NAS documents, images, audio and video for AI Q&A and chat over your files |
| AI search | Yes | Vendor confirmed | Natural language search over image content, voice content, documents, names |
| GPU expansion | Limited | Unverified | SFF-8654 PCIe 4.0 x8 external connector; wiki says external GPUs not recommended without official confirmation; no tested eGPU found |
| Hardware acceleration | Yes | Vendor confirmed | H.264 and H.265 8/10-bit codecs listed; ZettOS states hardware encoding and decoding |
| AI photos | Yes | Vendor confirmed | Photos app with People and Pets, scene grouping, duplicate detection |
| AI video | Yes | Vendor confirmed | Video and audio files are learned for content search; AI Transcription listed |
| Surveillance AI | Yes | Vendor confirmed | Surveillance Management app for ONVIF/RTSP IP cameras |
How much RAM does the D8 Ultra support?
Plan on upgrading. 32GB is fitted as standard, while the D8 Ultra accepts 96GB of DDR5 5600 SO-DIMM (dual channel), which is the difference between models up to the 14B class, with 32B class models a tight fit and 70B class models at 4-bit with room left for longer context (Estimated).
Our estimate for the biggest model that fits in 96GB is about 146B parameters at 4-bit, after setting aside 8GB for the system. Long context windows reduce that figure (Estimated).
Can you add a GPU to the D8 Ultra?
Yes, through a 1x external SFF-8654 connector (PCIe 4.0 x8); no internal PCIe card slot listed. That is the single biggest upgrade for local AI on the D8 Ultra: model layers move into fast graphics memory and replies speed up dramatically. Card size and power are the constraints.
The 11 TOPS NPU helps with photo recognition, transcription and vendor AI features. Most open-source runtimes such as Ollama do not use NAS NPUs yet, so do not count it towards language model speed.
Is the D8 Ultra suitable for RAG?
Yes. Retrieval augmented generation keeps your documents on the NAS and feeds the relevant passages to a local model. The D8 Ultra has the memory for an embedding model alongside a chat model, and 8 bays + 2 x M.2 for the library. Support is documented by the manufacturer.
The manufacturer also provides AI search across your files (documented by the manufacturer), which covers simple document questions without setting up your own stack.
For network access it has 2 x 10GbE. Fast enough that the network will not be the bottleneck.
Which D8 Ultra configurations are sold?
| Configuration | RAM | SSD | Notes |
|---|---|---|---|
| Without RAM | None | 256GB | Barebone, no memory; 256GB system SSD; pre-order only when checked |
| 32GB | 32GB | 256GB | 32GB DDR5 fitted; 256GB system SSD; pre-order only when checked |
How is the D8 Ultra different from the D6 Ultra?
The closest Zettlab model by AINAS Score. Only the specifications that differ are listed.
- Drive bays: 8 on the D8 Ultra, 6 on the D6 Ultra
- Power supply: none on the D8 Ultra, 240W on the D6 Ultra
- AINAS Score: 80 on the D8 Ultra, 78 on the D6 Ultra
Who should buy the D8 Ultra?
Best fit: home lab owners who want one box for storage, containers, media and a few AI services. That is where the D8 Ultra scores highest (Home Lab 79/100), followed by Local LLM (78/100).
Less suited to anyone who needs a system they can buy and receive today.
What are the alternatives to the D8 Ultra?
Other Native AI NAS systems closest in overall score.
D8 Ultra questions
Can the D8 Ultra run Ollama?
Partly. Ollama support is not yet verified. See the capability table above for the source.
Does Open WebUI work on the D8 Ultra?
Partly. Open WebUI support is not yet verified. See the capability table above for the source.
Does the D8 Ultra do AI photo recognition?
Yes. AI photos support is documented by the manufacturer. See the capability table above for the source.
Can the D8 Ultra run AI surveillance?
Yes. Surveillance AI support is documented by the manufacturer. See the capability table above for the source.
Does the D8 Ultra have hardware AI acceleration?
Yes. Hardware acceleration support is documented by the manufacturer. See the capability table above for the source.
How fast is the D8 Ultra's network?
It has 2 x 10GbE built in, and the 1x external SFF-8654 connector (PCIe 4.0 x8); no internal PCIe card slot listed can take a faster network card. The fastest port is 2 x 10GbE.
Where do these specifications come from?
Every field records its source. Specifications were last checked on . Spot an error? Send a correction with a manufacturer link.
- zettlab.com
- zettlab.com: ZettOS VM app; recommends D6/D8 Ultra for VMs
- wiki.zettlab.com: SFF-8654 x8 is the only GPU path; wiki says external GPUs are not recommended without official confirmation
- techradar.com: Manufacturer only publishes 34 TOPS combined CPU+GPU+NPU; TechRadar spec table lists Intel AI Boost NPU at 11 TOPS
- zettlab.com: Store shows Sold out and Pre-order item, shipping starts in March; all four variants unavailable in store JSON
- wiki.zettlab.com: ECC not stated; all validated modules are non-ECC DDR5 SO-DIMMs
- zettlab.com