16GB max RAM · 6 bays · 2 x 2.5GbE
AINAS Score 31
No verified price

AI CapableUGREEN
Why AI Capable: Intel N100 (4C/4T, no NPU) with UHD Graphics, 16GB max DDR5 and no GPU expansion path, so capable.
Updated
Can the UGREEN DXP4800 run local AI?
Partly. The UGREEN DXP4800 runs local models on Intel N100 alone, and its 16GB ceiling allows 7B to 8B class models (Estimated), but replies will be slow without graphics acceleration.
No graphics card can be fitted, which keeps every model on the CPU.
Local LLM support is documented by the manufacturer.
The DXP4800 suits buyers who want a four-bay home or small-office NAS that can also handle light AI jobs on the device itself. It uses an Intel N100 with four cores, comes with 8GB of DDR5 and takes up to 16GB. Two M.2 NVMe slots and two 2.5GbE ports cover fast storage and networking. UGOS Pro supports Docker and virtual machines, and UGREEN confirms an AI photo album for faces, places, objects and duplicates, plus Surveillance Center support. For language models, Qwen3 8B fits at Q4 (Estimated), as do Llama 3.2 3B and Llama 3.1 8B. The AINAS Score is 30, which reflects a system that is capable but limited. It comes in an 8GB diskless configuration with 32GB of eMMC system storage.
The limit that rules it out for some buyers is memory. The DXP4800 tops out at 16GB in a single DDR5 SO-DIMM slot, and it has no GPU path: no PCIe slot, no Thunderbolt and no OCuLink. Anyone who wants to run 14B models comfortably, or anything larger, will hit that wall. Qwen3 14B is only marginal at Q4 (Estimated), and Qwen3 32B does not fit at Q4 (Estimated). Those buyers should look for a memory ceiling higher than 16GB, plus a PCIe slot, OCuLink or Thunderbolt port so a graphics card can be added later. Buyers who only want file storage, photos and surveillance are not affected by this.
Amazon UK
8GB
£519.99£689.99
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 | Vendor confirmed | CPU-only Ollama in Docker per UGREEN KB; N100 and 16GB max limit this to small models. |
| Ollama | Yes | Vendor confirmed | UGREEN KB Docker Compose guide for Ollama/Open WebUI on UGOS Pro. |
| Open WebUI | Yes | Vendor confirmed | Official UGOS Pro Open WebUI guide. |
| RAG | Limited | Vendor confirmed | RAG via Open WebUI document chat only. |
| AI search | Limited | Vendor confirmed | Photos description and OCR search via Model Manager; document semantic search applicability per model not stated. |
| GPU expansion | No | Vendor confirmed | No PCIe slot, Thunderbolt or OCuLink. |
| Hardware acceleration | Limited | Vendor confirmed | Quick Sync on N100 UHD Graphics; no NPU. |
| AI photos | Yes | Vendor confirmed | AI photo album: faces, places, objects, duplicates. |
| AI video | Limited | Vendor confirmed | Surveillance Center local AI detection (people, vehicle, package, pet); features vary by model. |
| Surveillance AI | Yes | Vendor confirmed | Surveillance Center supports DXP series. |
Calculated from the DXP4800'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 | Q5 | Yes | 15GB | Not measured | Estimated |
| Qwen3 8BQwen, 8.19B | Q4 | Yes | 15GB | Not measured | Estimated |
| Gemma 3 12BGemma, 11.8B | Q4 | Marginal | 15GB | Not measured | Estimated |
| Qwen3 14BQwen, 14.8B | Q4 | Marginal | 16GB | Not measured | Estimated |
| Phi-4 14BPhi, 14.7B | Q4 | Marginal | 16GB | Not measured | Estimated |
| Mistral Small 3 24BMistral, 23.6B | None fits | No | 25GB at Q4 | Not measured | Estimated |
| Gemma 3 27BGemma, 27B | None fits | No | 29GB at Q4 | Not measured | Estimated |
| Qwen3 30B A3BQwen, 30.5B | None fits | No | 31GB at Q4 | Not measured | Estimated |
| Qwen3 32BQwen, 32.8B | None fits | No | 32GB at Q4 | Not measured | Estimated |
| DeepSeek-R1 32B (Qwen distill)DeepSeek, 32.8B | None fits | No | 32GB at Q4 | 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 DXP4800's use-case scores and capability rows. Unknown means we cannot confirm it, not that it probably works.
The DXP4800 is a poor choice for local LLM work. UGREEN confirms only limited support, with CPU-only Ollama in Docker, and says the N100 and 16GB maximum restrict it to small models.
Private RAG is a poor fit here. Ollama and Open WebUI are vendor confirmed on UGOS Pro, but RAG support is limited to document chat in Open WebUI, running on the same CPU-only hardware.
AI search is rated poor on the DXP4800. UGREEN confirms limited support through photo description and OCR search in Model Manager, and does not state whether document semantic search applies on each model.
AI photos is rated poor, even though the feature itself is present. UGREEN confirms an AI photo album covering faces, places, objects and duplicates, but the system has no NPU and only the N100's integrated graphics.
Surveillance is rated poor for this system. Surveillance Center supports the DXP series, and local AI detection of people, vehicles, packages and pets is limited, with features varying by model.
Home lab use is rated poor. Docker and virtual machines are both supported on UGOS Pro, but the four-core N100, 16GB memory ceiling and absence of ECC memory leave little room for heavier workloads.
Business use is rated poor. The DXP4800 has two 2.5GbE ports, four bays and a two-year warranty, but no ECC memory, a single memory slot and no GPU or PCIe expansion for growth.
Creative AI is a poor fit. GPU expansion is confirmed as not supported, with no PCIe slot, Thunderbolt or OCuLink, and hardware acceleration is limited to Quick Sync on the N100 graphics.
Out of the box the DXP4800 has 8GB; the official ceiling is 16GB of DDR5 SO-DIMM 4800. A memory upgrade is the cheapest way to make it more useful for AI.
Working backwards from 16GB: 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 13B (Estimated).
There is no supported way to add a GPU to the DXP4800: no PCIe slot and no OCuLink. Inference runs on the CPU, with Intel UHD Graphics handling media tasks, so keep to small models.
For a personal library, yes. A compact embedding model and a small chat model will run, but the DXP4800 is better as the document store behind a RAG system whose model runs on stronger hardware.
Built-in networking: 2 x 2.5GbE. Adequate for a household, slower for bulk indexing.
| Configuration | RAM | SSD | Notes |
|---|---|---|---|
| 8GB | 8GB | None | Diskless, 8GB DDR5, 32GB eMMC system storage. |
The closest UGREEN model by bay count and AINAS Score. Only the specifications that differ are listed.
Its strongest AINAS profiles are Home Lab at 33 and Business at 33, which points to home lab owners who want one box for storage, containers, media and a few AI services.
Look elsewhere if you are anyone who wants to run models larger than about 8B parameters, or buyers who expect to add a graphics card later.
Other AI Capable systems closest in price.
Yes, at Q4. Qwen3 8B needs 5.2GB at Q4 and fits (Estimated). At Q8 it needs 8.9GB, which is marginal (Estimated), meaning it loads with little headroom. UGREEN documents CPU-only Ollama in Docker and says the N100 and 16GB maximum limit this to small models. Qwen3 14B is marginal at Q4 (Estimated), so Qwen3 8B is the largest model listed on this page that fits comfortably.
Yes. The DXP4800 has one DDR5 SO-DIMM 4800 slot and ships with 8GB. The maximum is 16GB, so you can double the memory but go no further. The memory is not ECC. The 16GB ceiling is the main constraint on which language models the system can hold, so check model sizes before buying.
No. UGREEN confirms there is no PCIe slot, no Thunderbolt and no OCuLink, and the page lists zero USB4 ports. The only graphics hardware is the Intel UHD Graphics inside the N100, which offers Quick Sync acceleration. If you want to add a GPU for larger models or creative AI, you need a system with a PCIe slot or an OCuLink or Thunderbolt connection.
Yes. UGOS Pro on the DXP4800 supports Docker and virtual machines, and local LLM support is limited to CPU-only Ollama in Docker. The four-core Intel N100 and the 8GB fitted suit a few containers or one light virtual machine, and the 16GB ceiling in a single DDR5 SO-DIMM slot caps how far that can grow. Quick Sync on the UHD Graphics handles video transcoding for media apps running alongside.
Yes. It has two M.2 2280 NVMe slots, alongside four drive bays. That gives you room for fast storage next to the main drives. The M.2 slots do not add any AI capability, since the system has no NPU and no GPU expansion, but they suit caching or fast storage for Docker containers.
The DXP4800 has a two-year warranty, listed by UGREEN as two years of after-sales support. The page does not give further terms, so check with UGREEN for what the cover includes. The system was released in 2024 and runs UGOS Pro, with Docker and virtual machine support.
Yes. Ollama support is documented by the manufacturer. See the capability table above for the source.
Yes. Open WebUI support is documented by the manufacturer. 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.
Yes. Surveillance AI support is documented by the manufacturer. 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.