96GB max RAM · 4 bays · 2 x 10GbE
AINAS Score 81
No verified UK price
View
Can the QNAP QAI-h1290FX run local AI?
Yes, the QNAP QAI-h1290FX can run local language models. It pairs AMD EPYC 7302P with up to 1024GB of memory, which suits 70B class models at higher-quality quantization, or large mixture-of-experts models at 4-bit (Estimated).
It also has a GPU path (Full-size GPU), so a graphics card can be added later for much faster inference.
Local LLM support is documented by the manufacturer.
What is the QAI-h1290FX like?
The QAI-h1290FX is the most powerful system we track and the only one built as a full AI server rather than a NAS with AI features. It pairs a 16-core AMD EPYC 7302P with 128GB of ECC memory that expands to 1TB, twelve U.2 NVMe bays, two 25GbE ports and four PCIe Gen4 slots that take full-size workstation GPUs.
QNAP sells it with or without a graphics card, and the UK bundles use NVIDIA RTX PRO Blackwell cards. QNAP's launch material says Ollama, Open WebUI and AnythingLLM come preloaded, and its Qsirch search can run in local AI mode with a 16GB or larger GPU. That makes it the clearest vendor-confirmed local LLM and RAG platform we track.
This is business hardware with business pricing and a five-year warranty. It suits teams that want a private model serving several users, or a RAG system over company documents that never leaves the building. For a home user it is far beyond what is needed; the TS-h1290FX is the same chassis without the AI bundle.
Current UK offers
Not currently on Amazon UK
We could not find a UK listing for the QNAP QAI-h1290FX 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
- 4x PCIe Gen4 (slot 1 x16, slot 2 x16, slot 3 x8, slot 4 x16); slots 3 and 4 take cards up to 280 mm, wider cards if next slot unused
- Source: Manufacturer
- Docker
- Yes
- VM support
- Yes
- Released
- 2026
Can the QAI-h1290FX 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 | Product page: local deployment of private LLMs and RAG; supports vLLM inference engine. |
| Ollama | Yes | Vendor confirmed | Press release: preloaded AI tools include AnythingLLM, OpenWebUI and Ollama. |
| Open WebUI | Yes | Vendor confirmed | OpenWebUI listed among preloaded AI tools. |
| RAG | Yes | Vendor confirmed | Product page: private LLMs and RAG-based search; Qsirch RAG search also available. |
| AI search | Yes | Vendor confirmed | Qsirch page lists QAI-h1290FX as AI Mode Ready (local inference, needs 16GB+ VRAM GPU). |
| GPU expansion | Yes | Vendor confirmed | 4 PCIe Gen4 slots; compatibility list includes RTX PRO 6000 Blackwell Max-Q (96GB) and RTX PRO 4500 Blackwell (32GB). |
| Hardware acceleration | Limited | Vendor confirmed | No integrated GPU or NPU; acceleration depends on the NVIDIA GPU fitted (bundled in UK B4500/B6000 SKUs). |
| AI photos | Yes | Vendor confirmed | QuMagie AI-assisted photo management (people, things, places) listed in software specs. |
| AI video | Limited | Vendor confirmed | QVR Face listed only under QTS; default OS is QuTS hero. |
| Surveillance AI | Yes | Vendor confirmed | QVR Surveillance: 2 free channels, register to claim 6 more; QVR Pro 8 free channels on QTS. |
How much RAM does the QAI-h1290FX support?
The QAI-h1290FX ships with 128GB of DDR4 ECC RDIMM and supports up to 1024GB with ECC. Upgrading to the maximum lifts the model ceiling from 70B class models at higher-quality quantization, or large mixture-of-experts models at 4-bit to 70B class models at higher-quality quantization, or large mixture-of-experts models at 4-bit (both Estimated).
As a rule of thumb, a 4-bit model needs about 0.6GB per billion parameters plus room for context. After leaving around 8GB for the NAS operating system, 1024GB leaves space for a model of roughly 1693B parameters at most (Estimated).
ECC memory support is a plus for always-on systems that also hold important files.
Can you add a GPU to the QAI-h1290FX?
Yes. The QAI-h1290FX has a 4x PCIe Gen4 (slot 1 x16, slot 2 x16, slot 3 x8, slot 4 x16); slots 3 and 4 take cards up to 280 mm, wider cards if next slot unused that can take a graphics card. A discrete GPU with its own memory runs models far faster than CPU inference and lets the NAS serve several users at once. Check card length, height and power draw against the chassis and the 750W power supply first.
Is the QAI-h1290FX suitable for RAG?
Yes. Retrieval augmented generation keeps your documents on the NAS and feeds the relevant passages to a local model. The QAI-h1290FX has the memory for an embedding model alongside a chat model, and 12 bays 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.
Networking is 2 x 25GbE + 2 x 2.5GbE. That is ample for indexing large libraries and for several people querying at once.
Which QAI-h1290FX configurations are sold?
| Configuration | RAM | SSD | Notes |
|---|---|---|---|
| QAI-h1290FX-B6000 | None | None | UK bundle; SKU name implies RTX PRO 6000 Blackwell Max-Q (96GB), not stated explicitly on the page |
| QAI-h1290FX-7302P-128G (no GPU) | 128GB | None | Only configuration on the QNAP spec page; GPU optional. Not listed on UK store. |
| QAI-h1290FX-B4500-UK | 128GB | None | UK bundle with NVIDIA RTX PRO 4500 Blackwell (32GB GDDR7). SSD inclusion not stated. |
How is the QAI-h1290FX different from the TS-h1290FX?
The closest QNAP model by AINAS Score. Only the specifications that differ are listed.
- Processor: AMD EPYC 7302P on the QAI-h1290FX, AMD EPYC 7232P (or 7252) on 64G model; EPYC 7302P on 128G/256G models on the TS-h1290FX
- CPU cores: 16 on the QAI-h1290FX, 8 on the TS-h1290FX
- Memory fitted: 128GB on the QAI-h1290FX, 64GB on the TS-h1290FX
- AI class: Native AI NAS on the QAI-h1290FX, AI Ready on the TS-h1290FX
- AINAS Score: 86 on the QAI-h1290FX, 75 on the TS-h1290FX
Who should buy the QAI-h1290FX?
The QAI-h1290FX scores best for Home Lab (90/100) and RAG (88/100). It suits home lab owners who want one box for storage, containers, media and a few AI services.
What are the alternatives to the QAI-h1290FX?
Other Native AI NAS systems closest in overall score.
QAI-h1290FX questions
Can the QAI-h1290FX run Ollama?
Yes. Ollama support is documented by the manufacturer. See the capability table above for the source.
Does Open WebUI work on the QAI-h1290FX?
Yes. Open WebUI support is documented by the manufacturer. See the capability table above for the source.
Does the QAI-h1290FX do AI photo recognition?
Yes. AI photos support is documented by the manufacturer. See the capability table above for the source.
Can the QAI-h1290FX run AI surveillance?
Yes. Surveillance AI support is documented by the manufacturer. See the capability table above for the source.
Does the QAI-h1290FX have hardware AI acceleration?
Partly. Hardware acceleration support is documented by the manufacturer. See the capability table above for the source.
How fast is the QAI-h1290FX's network?
It has 2 x 25GbE + 2 x 2.5GbE built in, and the 4x PCIe Gen4 (slot 1 x16, slot 2 x16, slot 3 x8, slot 4 x16); slots 3 and 4 take cards up to 280 mm, wider cards if next slot unused can take a faster network card. The fastest port is 25GbE.
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.
- qnap.com: 3 x USB 3.2 Gen 1 only
- ukstore.qnap.com
- qnap.com
- qnap.com: GPGPU list includes NVIDIA RTX PRO 6000 Blackwell Max-Q, RTX PRO 4500 Blackwell, RTX PRO 4000 Blackwell, RTX 4000 Ada, RTX A4000
- qnap.com: Press release dated 30 April 2026
- qnap.com: Product page: 5 year warranty; spec page lists bare value 5