64GB max RAM · 4 bays · 2 x 10GbE
AINAS Score 50
No verified price

AI ReadyAoostar
Why AI Ready: Ryzen 7 5825U (no NPU) with Radeon iGPU, no PCIe/OCuLink/USB4 GPU path, but Aoostar states up to 64GB DDR4, which meets the ready memory threshold; borderline, could be argued capable as the CPU is a 2022 Zen 3 part.
Updated
Can the Aoostar WTR Pro run local AI?
The Aoostar WTR Pro can run small local models, with limits. Its AMD Ryzen 7 5825U and 64GB memory ceiling suit 32B class models comfortably, and 70B class models at 4-bit with little headroom (Estimated), so treat it as a storage-first NAS that can host light AI tools.
There is no way to add a discrete GPU, so all inference runs on the CPU and basic integrated graphics.
Local LLM support is confirmed by community reports.
The WTR Pro suits a buyer who wants a 4-bay compact desktop that doubles as a self-managed local AI box. It takes up to 64GB of DDR4 3200 SO-DIMM memory across two slots, and it has an AMD Ryzen 7 5825U with 8 cores, 2 x 2.5GbE and 2 M.2 slots. Docker and virtual machines are both supported. Llama 3.3 70B fits at Q4 (Estimated), and every model up to 32B fits at both Q4 and Q8 (Estimated). It comes as a barebone with a Windows 11 Pro key, so you add the memory and drives yourself. Ollama has a native Windows build, and that route is community confirmed.
The main thing that rules it out is the lack of any GPU path. The WTR Pro has no PCIe slot, no OCuLink and no USB4, and the Radeon iGPU is the only graphics hardware, with no NPU. Inference runs on the CPU only, which the capability notes describe as slow. Buyers who want responsive large models, or who plan to add a graphics card later, should look for a system with OCuLink, USB4 or a PCIe slot, or a machine with an NPU. It also has no vendor NAS software, so you build the storage layer yourself.
Amazon US
Barebone
Ships from the US; delivery and import charges shown at checkout
$399.00
Verified
In stock
Buy on AmazonAmazon US
Barebone
Ships from the US; delivery and import charges shown at checkout
$399.00
Verified
In stock
Buy on AmazonProduct 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 or Amazon.com, as applicable, 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 | Limited | Community confirmed | No vendor AI software. Ships with Windows 11; Ollama has a native Windows build. CPU-only inference; up to 64GB DDR4 allows mid-size quantised models slowly. |
| Ollama | Yes | Community confirmed | Ollama offers a native Windows installer; Linux NAS OS installs are user choice. |
| Open WebUI | Yes | Unverified | Runnable via Docker or Python on Windows/Linux; no platform specific source found. |
| RAG | Limited | Unverified | Only via user installed tools such as Open WebUI; no vendor feature. |
| AI search | Limited | Unverified | No vendor NAS OS or AI search; third party apps only. |
| GPU expansion | No | Vendor confirmed | No OCuLink, USB4 or PCIe slot. |
| Hardware acceleration | Limited | Vendor confirmed | Radeon iGPU only; no NPU. |
| AI photos | Limited | Unverified | No vendor photo app; third party apps such as Immich via Docker not verified on this hardware. |
| AI video | Limited | Unverified | No vendor feature; third party only, not verified. |
| Surveillance AI | Limited | Unverified | No vendor surveillance app; third party software only, not verified. |
Calculated from the WTR Pro's memory and each model's published size at each quantization. Rows marked Estimated are calculations, not benchmarks. Check any model.
| Model | Best quantization | Runs? | Memory needed | Expected performance | Verification |
|---|---|---|---|---|---|
| Llama 3.2 3BLlama, 3.21B | Q8 | Yes | 13GB | Not measured | Estimated |
| Llama 3.1 8BLlama, 8.03B | Q8 | Yes | 19GB | Not measured | Estimated |
| Qwen3 8BQwen, 8.19B | Q8 | Yes | 19GB | Not measured | Estimated |
| Gemma 3 12BGemma, 11.8B | Q8 | Yes | 24GB | Not measured | Estimated |
| Qwen3 14BQwen, 14.8B | Q8 | Yes | 28GB | Not measured | Estimated |
| Phi-4 14BPhi, 14.7B | Q8 | Yes | 28GB | Not measured | Estimated |
| Mistral Small 3 24BMistral, 23.6B | Q8 | Yes | 38GB | Not measured | Estimated |
| Gemma 3 27BGemma, 27B | Q8 | Yes | 44GB | Not measured | Estimated |
| Qwen3 30B A3BQwen, 30.5B | Q8 | Yes | 47GB | Not measured | Estimated |
| Qwen3 32BQwen, 32.8B | Q8 | Yes | 50GB | Not measured | Estimated |
| DeepSeek-R1 32B (Qwen distill)DeepSeek, 32.8B | Q8 | Yes | 50GB | Not measured | Estimated |
| Llama 3.3 70BLlama, 70.6B | Q4 | Yes | 60GB | Not measured | Estimated |
| Llama 4 Scout 17Bx16ELlama, 109B | None fits | No | 89GB at Q4 | Not measured | Estimated |
| Qwen2.5 72BQwen, 72.7B | Q3 | Yes | 54GB | Not measured | Estimated |
| Mistral Large 2 123BMistral, 123B | Q2 | Yes | 62GB | Not measured | Estimated |
One verdict per workload, from the WTR Pro's use-case scores and capability rows. Unknown means we cannot confirm it, not that it probably works.
The WTR Pro is an OK choice for local LLMs. Support is limited and community confirmed, running CPU-only through Ollama on Windows, with up to 64GB of DDR4 allowing mid-size quantised models slowly.
Private RAG is an OK fit on the WTR Pro. It depends on user-installed tools such as Open WebUI, which is unverified here, and there is no vendor RAG feature, though Ollama is community confirmed.
We cannot confirm AI search on the WTR Pro. The capability is listed as limited and unverified, because there is no vendor NAS OS or search feature and only third party apps would apply.
We cannot confirm AI photo features on the WTR Pro. There is no vendor photo app, and third party apps such as Immich via Docker are unverified on this hardware, even though Docker is supported.
Surveillance is unconfirmed on the WTR Pro. It has no vendor surveillance app, and third party software is unverified, so we cannot say how well camera analysis would work.
The WTR Pro is OK for a home lab. It supports Docker and virtual machines, has 8 cores, takes up to 64GB of memory, and offers 2 x 2.5GbE with 4 drive bays and 2 M.2 slots.
The WTR Pro is OK for business use. It ships with a Windows 11 Pro key, supports virtual machines, and offers 4 bays and 2 x 2.5GbE, though it has no ECC memory and a 1 year warranty.
Creative AI is OK on the WTR Pro, with limits. The vendor confirms there is no GPU expansion, and hardware acceleration is limited to the Radeon iGPU, so work stays CPU-bound and slow.
Sold barebone or with memory bundles, the WTR Pro accepts 64GB of DDR4 3200 SO-DIMM. Buy as much as your budget allows: memory decides model size.
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, 64GB leaves space for a model of roughly 93B parameters at most (Estimated).
No. The WTR Pro has no PCIe slot or OCuLink port for a graphics card. Inference runs on the CPU, with AMD Radeon Graphics handling media tasks, so keep to small models.
With limits. The WTR Pro can host a small RAG stack such as AnythingLLM or Open WebUI with a compact embedding model, but memory and compute restrict how large the answering model can be. It suits a personal document library more than a team knowledge base.
Networking is 2 x 2.5GbE. Fine for one or two users; indexing a large archive over the network will be slower.
| Configuration | RAM | SSD | Notes |
|---|---|---|---|
| Barebone | None | None | Free Windows 11 Pro key. |
The closest Aoostar model by bay count and AINAS Score. Only the specifications that differ are listed.
The WTR Pro scores best for Business (53/100) and Local LLM (51/100). It suits small offices that need shared storage with private AI tools and room to grow.
It is a weaker fit for buyers who expect to add a graphics card later and for teams moving large files over the network every day.
Other AI Ready systems closest in price.
Buying guides that list it
Yes at Q4, as an estimate. Llama 3.3 70B needs 43GB at Q4 and the WTR Pro takes up to 64GB, so it fits (Estimated). At Q8 it does not fit (Estimated). The capability notes say CPU-only inference is slow, so expect it to be slow in practice.
Up to 64GB. It uses two DDR4 3200 SO-DIMM slots, and the memory is expandable. The system ships with 0GB included, so you supply the memory. There is no ECC support.
No. The vendor confirms there is no PCIe slot, no OCuLink and no USB4. The Radeon integrated graphics are the only graphics hardware, and there is no NPU.
Yes, two. Both are M.2 2280 NVMe slots at PCIe 3.0, and they sit alongside the 4 SATA bays. Aoostar sells one configuration, the barebone with a Windows 11 Pro key, and it includes no drives, so you supply the NVMe and SATA storage yourself along with the memory. Two fast slots suit an operating system drive plus an app or cache drive, which keeps all four bays free for hard drives.
Yes. Docker and virtual machines are both supported. Open WebUI can run via Docker or Python, but that is unverified on this system. Ollama has a native Windows build, which is community confirmed.
Windows 11 Pro. A key is included, and there is no Aoostar NAS OS. The vendor page says to install your preferred system, and Linux installs are the user's choice.
Yes. Ollama support is confirmed by community reports. See the capability table above for the source.
Yes. Open WebUI support is not yet verified. See the capability table above for the source.
Partly. AI photos support is not yet verified. See the capability table above for the source.
Partly. Surveillance AI support is not yet verified. See the capability table above for the source.
Partly. Hardware acceleration support is documented by the manufacturer. See the capability table above for the source.
It has 2 x 2.5GbE built in. The fastest port is 2 x 2.5GbE.
Every field records its source. Specifications were last checked on . Spot an error? Send a correction with a manufacturer link.