12GB max RAM · 4 x M.2 · 2 x 2.5GbE
AINAS Score 28
No verified price

AI CapableSynologyDiscontinued
Why AI Capable: AMD Ryzen V1500B with no NPU and no iGPU, 32 GB maximum memory, and a PCIe Gen3 x8 (x4 link) slot Synology supports only for NICs and M.2 adapters, so CPU-only small models via Container Manager.
Updated
Can the Synology RS822+ run local AI?
The Synology RS822+ can run small local models, with limits. Its AMD Ryzen V1500B 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 RS822+ suits a buyer who wants a 4-bay rackmount Synology NAS for storage, Surveillance Station and containers, with AI as a minor extra. It has an AMD Ryzen V1500B with 4 cores, DDR4 ECC SODIMM memory that expands to 32GB, four 1GbE ports, and Docker and Virtual Machine Manager on DSM. Surveillance Station supports up to 40 IP cameras at 1080p H.264, with 2 free licences, and the warranty is 3 years. AI runs on the CPU only. Ollama and Open WebUI run in containers and are community confirmed, but the base 2GB of memory must be upgraded first. It has an AINAS Score of 28, which reflects that limited AI role.
The limit that rules it out for AI buyers is the lack of any GPU or NPU path. The Ryzen V1500B has no integrated GPU and no NPU, and Synology lists the PCIe Gen3 x8 slot (x4 link) only for 10GbE and 25GbE network cards and M.2 adapters. No GPU support is documented, and the chassis is 1U. Anyone planning regular local LLM use, private document chat or Synology AI Search should look for a system with a supported GPU or an NPU with a stated TOPS figure, plus a memory ceiling above 32GB. Synology has also discontinued the RS822+, and its memory starts at only 2GB, which has to be raised before any model work begins.
Amazon UK
RS822+ 2GB ECC
No current price
Checked , no price yet
Out of 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 | CPU-only Ollama via Container Manager; guide uses a V1500B DS925+ with small (1B to 3B) models. Base 2 GB RAM must be upgraded. |
| Ollama | Yes | Community confirmed | Not in Package Center; ollama/ollama container runs in Container Manager on DSM (CPU only). |
| Open WebUI | Yes | Community confirmed | Open WebUI container documented on Synology DSM pointing at Ollama. |
| RAG | Limited | Unverified | No Synology local RAG on this model; AI Console bridges to cloud or OpenAI-compatible APIs. Open WebUI document chat via Docker plausible but not verified on this unit. |
| AI search | No | Vendor confirmed | Synology AI Search in Drive needs GPU-powered models, available only on supported systems with a GPU (AI Console 2.0 beta lists FS3420, FS6420, RS11626xs+, RS3626xs, RS4826xs+, RS6426xs+); this model is not listed. |
| GPU expansion | No | Vendor confirmed | PCIe Gen3 x8 (x4 link) slot exists but Synology lists only 10/25GbE NICs and M.2 adapter cards; 1U chassis; no GPU support documented. |
| Hardware acceleration | No | Community confirmed | Ryzen V1500B has no integrated GPU and no NPU; no hardware transcoding or AI acceleration. |
| AI photos | Limited | Vendor confirmed | Synology Photos facial and object recognition listed; object recognition needs a memory upgrade from the 2 GB default. |
| AI video | Limited | Community confirmed | No on-NAS deep video analytics; Surveillance Station can use on-camera AI from Synology BC500/TC500 cameras. |
| Surveillance AI | Yes | Vendor confirmed | Surveillance Station up to 40 IP cameras (1080p H.264); 2 free licences. |
Calculated from the RS822+'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 RS822+'s use-case scores and capability rows. Unknown means we cannot confirm it, not that it probably works.
The RS822+ is a poor choice for a local LLM. Support is limited and community confirmed, running CPU-only Ollama in Container Manager, and the guide used small 1B to 3B models, while the base 2GB of memory must be upgraded first.
Private RAG is poor here. The rag row is limited and Unverified: Synology offers no local RAG on this model, and Open WebUI document chat through Docker is plausible but not verified on this unit, so treat it as unconfirmed.
AI search is poor, because the RS822+ does not support it. Synology AI Search in Drive needs GPU-powered models on supported GPU systems, and the RS822+ is not on the vendor's list, which is vendor confirmed.
AI photos is poor, with only partial support. Synology Photos lists facial and object recognition, vendor confirmed as limited, and object recognition needs a memory upgrade from the 2GB default, with no GPU or NPU to help.
Surveillance is rated poor for AI work, despite basic recording support. Surveillance Station is vendor confirmed for up to 40 IP cameras at 1080p H.264, but there is no on-NAS deep video analytics, only on-camera AI from Synology cameras.
Home lab is poor on this rating. DSM runs Docker and Virtual Machine Manager, and memory expands to 32GB, but the CPU has 4 cores, there are four 1GbE ports, no M.2 slots, and only 2GB of memory to start.
Business use is poor on this rating. It has ECC memory, four 1GbE ports and a 3-year warranty, but Synology has discontinued it, which makes long-term planning harder, and it ships with only 2GB of memory.
Creative AI is poor. GPU expansion is not supported, vendor confirmed, and there is no hardware acceleration, since the Ryzen V1500B has no integrated GPU and no NPU, so GPU-based creative tools have no route here.
The RS822+ ships with 2GB of DDR4 ECC SODIMM and supports up to 32GB with ECC. Upgrading to the maximum lifts the model ceiling from only very small models of around 1B to 3B parameters to models up to the 14B class, with 32B class models a tight fit (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, 32GB leaves space for a model of roughly 40B parameters at most (Estimated).
ECC support protects against silent memory errors on a box that runs around the clock.
No. The RS822+ has no PCIe slot or OCuLink port for a graphics card. Inference runs on the CPU, so keep to small models.
With limits. The RS822+ 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 4 x 1GbE. Fine for one or two users; indexing a large archive over the network will be slower.
| Configuration | RAM | SSD | Notes |
|---|---|---|---|
| RS822+ 2GB ECC | 2GB | None | 2 GB DDR4 ECC SODIMM (1x 2 GB), 2 slots, max 32 GB. |
The closest Synology model by bay count and AINAS Score. Only the specifications that differ are listed.
The RS822+ scores best for Home Lab (30/100) and Business (30/100). It suits home lab owners who want one box for storage, containers, media and a few AI services.
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, on paper. DeepSeek-R1 32B (Qwen distill) fits at Q4 (Estimated), needing 20GB against the 32GB maximum, but it does not fit at Q8 (Estimated), which needs 35GB. This is a calculation, not a benchmark, and it assumes the memory upgrade from the 2GB included. Running is CPU only, and the community confirmed Ollama guide covers small 1B to 3B models.
Yes. It ships with 2GB of DDR4 ECC SODIMM as one module and takes up to 32GB, listed as 16GB x 2 across 2 slots. The upgrade matters for AI use: the Ollama guide says the base 2GB must be upgraded, and Synology Photos object recognition also needs more than the default.
Yes. The RS822+ uses DDR4 ECC SODIMM, and the 2GB module it ships with is ECC. The configuration sold is the RS822+ 2GB ECC, with 2 memory slots and a 32GB maximum. ECC protects stored data in memory from single-bit errors, which suits a NAS that runs around the clock.
No. The PCIe Gen3 x8 slot runs as a x4 link, and Synology lists only 10GbE and 25GbE network cards and M.2 adapters for it. GPU support is vendor confirmed as not documented, and the chassis is 1U. There is also no OCuLink, Thunderbolt or USB4 port for an external GPU.
Yes. The RS822+ runs Synology DSM with Docker containers and virtual machines both supported. The catch is memory: the RS822+ 2GB ECC configuration has only 2GB, so containers such as Ollama and Open WebUI, which are community confirmed on DSM, need an upgrade first. The 2 slots take DDR4 ECC SODIMM up to 32GB, which also lifts model fit as far as DeepSeek-R1 32B at Q4 (Estimated).
No. It has 0 M.2 slots. NVMe cache is possible only through an add-in card in the PCIe slot, and Synology lists the M2D20, M2D18 and E10M20-T1 for that. The slot is a single PCIe Gen3 x8 with a x4 link, so a cache card and a faster network card cannot both be fitted.
Yes. Ollama support is confirmed by community reports. See the capability table above for the source.
Yes. Open WebUI support is confirmed by community reports. See the capability table above for the source.
Partly. 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.
No. Hardware acceleration support is confirmed by community reports. See the capability table above for the source.
It has 4 x 1GbE built in, and the 1x PCIe Gen3 x8 slot (x4 link); Synology-listed cards: E10G30-F2/E10G30-T2/E10G18-T1 (10GbE), E25G30-F2 (25GbE), E10M20-T1, M2D20/M2D18 can take a faster network card. The fastest port is 1GbE.
Every field records its source. Specifications were last checked on . Spot an error? Send a correction with a manufacturer link.