16GB max RAM · 2 bays · 2.5GbE
AINAS Score 29
No verified price

AI CapableAoostar
Why AI Capable: Intel N150 (no NPU), single DDR4 SO-DIMM up to 32GB, no GPU expansion path; Docker capable x86 at best.
Updated
Can the Aoostar R1 run local AI?
The Aoostar R1 can run small local models, with limits. Its Intel N150 and 32GB memory ceiling suit models up to the 14B class, with 32B class models a tight fit (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.
Local LLM support is confirmed by community reports.
The R1 suits a buyer who wants a small, low-cost x86 box for light file serving and Docker, and is happy to install and manage the software themselves. It has an Intel N150 with 4 cores, two SATA bays, one M.2 slot and two 2.5GbE ports. It supports Docker and virtual machines. One DDR4 3200 SO-DIMM slot takes up to 32GB, and that slot can be upgraded. Ollama has a native Windows build and is community confirmed. Qwen3 14B fits at Q4 and Q8 (Estimated), and DeepSeek-R1 32B fits at Q4 (Estimated). The AINAS Score is 29.
The main problem is that the R1 has no route to a GPU. It has no PCIe slot, no OCuLink and no USB4, and the vendor confirms there is no GPU expansion. It also has no NPU. Anyone who wants local AI beyond small models should look for a system with a PCIe slot, OCuLink or USB4 for a graphics card, or an NPU with a stated TOPS figure. Larger memory helps too. The R1 tops out at 32GB, so Llama 3.3 70B does not fit at Q4 (Estimated). There is no ECC memory support either, and the warranty is 1 year, so it suits a hobby server better than business file storage.
No verified price
We have not verified a listing for the Aoostar R1 at a retailer we track. The specifications below stand on their own; a price appears here as soon as a 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 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. Small models only with up to 32GB. |
| 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 | Unverified | Intel N150 iGPU not listed by Aoostar; 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 R1'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 | Q4 | Yes | 25GB | Not measured | Estimated |
| Gemma 3 27BGemma, 27B | Q4 | Yes | 29GB | Not measured | Estimated |
| Qwen3 30B A3BQwen, 30.5B | Q4 | Yes | 31GB | Not measured | Estimated |
| Qwen3 32BQwen, 32.8B | Q4 | Yes | 32GB | Not measured | Estimated |
| DeepSeek-R1 32B (Qwen distill)DeepSeek, 32.8B | Q4 | Yes | 32GB | Not measured | Estimated |
| Llama 3.3 70BLlama, 70.6B | None fits | No | 60GB at Q4 | Not measured | Estimated |
| Llama 4 Scout 17Bx16ELlama, 109B | None fits | No | 89GB at Q4 | Not measured | Estimated |
| Qwen2.5 72BQwen, 72.7B | None fits | No | 65GB at Q4 | Not measured | Estimated |
| Mistral Large 2 123BMistral, 123B | None fits | No | 96GB at Q4 | Not measured | Estimated |
One verdict per workload, from the R1's use-case scores and capability rows. Unknown means we cannot confirm it, not that it probably works.
The R1 is a poor choice for local LLM work. Support is limited and community confirmed, with Ollama on Windows and small models only, and the 32GB memory ceiling with no GPU path or NPU holds it back.
The R1 is a poor choice for private RAG. RAG is limited and Unverified, possible only through tools you install yourself such as Open WebUI, which is also Unverified here, and there is no vendor feature.
We cannot confirm AI search on the R1. The capability is limited and Unverified, because there is no vendor NAS OS or search feature and only third-party apps would be an option.
We cannot confirm AI photo features on the R1. There is no vendor photo app, and third-party tools such as Immich through Docker are not verified on this hardware.
We cannot confirm surveillance use on the R1. It has no vendor surveillance app, and any third-party software is unverified, so camera analysis is not a confirmed feature.
The R1 is a poor home lab fit. It supports Docker and virtual machines, but a single SO-DIMM slot capped at 32GB and no PCIe or fast networking beyond 2.5GbE limit what you can run.
The R1 is a poor business choice. It has no ECC memory, SATA bays that are not hot-swap, a one-year warranty and no vendor NAS software, so it lacks the safeguards a business would expect.
The R1 is a poor choice for creative AI. GPU expansion is confirmed absent by the vendor, and hardware acceleration is limited and Unverified because Aoostar does not list the N150 graphics and there is no NPU.
The R1 can be bought without memory and takes up to 32GB of DDR4 3200 SO-DIMM (1 slot). Fitting the maximum is what turns it into a serious local AI machine.
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, 32GB leaves space for a model of roughly 40B parameters at most (Estimated).
No. The R1 has no PCIe slot or OCuLink port for a graphics card. Inference runs on the CPU, so keep to small models.
With limits. The R1 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 | No RAM, no SSD, no OS. |
| 16GB + 512GB | 16GB | 512GB | |
| 32GB + 1TB | 32GB | 1.024TB |
The closest Aoostar model by bay count and AINAS Score. Only the specifications that differ are listed.
The R1 scores best for Business (32/100) and Home Lab (31/100). 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 Capable systems closest in overall score.
Yes, at Q4 only. DeepSeek-R1 32B (Qwen distill) needs 20GB at Q4 and 35GB at Q8, so it fits at Q4 and not at Q8 (Estimated). That is a calculation, not a benchmark, and the 32GB ceiling leaves little headroom. Ollama is community confirmed on Windows, but local LLM support is limited and suits small models best.
Yes. The R1 has one DDR4 3200 SO-DIMM slot that takes up to 32GB. The barebone version ships with no memory, so you supply it. Other configurations are listed with 16GB or 32GB. With only one slot, 32GB is the maximum, and the system has no ECC memory support.
No. The vendor confirms there is no OCuLink, USB4 or PCIe slot, so there is no way to attach a GPU. The R1 has no NPU either. Aoostar does not list the Intel N150 graphics, and hardware acceleration is limited and Unverified.
No. Estimated: Llama 3.3 70B is 43GB at Q4 and 75GB at Q8, and with a 32GB memory ceiling neither level fits on the R1. The same is true of every larger model on this page, including Qwen2.5 72B and Llama 4 Scout 17Bx16E (Estimated). The largest model that fits is DeepSeek-R1 32B (Qwen distill) at Q4 (Estimated). This is a calculation, not a benchmark, and inference runs on the CPU because there is no GPU path or NPU.
Three configurations are listed. The barebone has no RAM, SSD or operating system. A 16GB and 512GB version and a 32GB and 1TB version also appear on the archived vendor page, both with Windows 11 Pro preinstalled. Every version uses the same single DDR4 3200 SO-DIMM slot, so the 32GB version already fills the memory ceiling.
Windows 11 Pro is preinstalled on configured units. Aoostar has not developed NAS software, so there is no Aoostar NAS OS. The barebone has no operating system or key. Linux NAS OS installs are the buyer's choice, and the R1 supports Docker and virtual machines.
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 not yet verified. 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.