96GB max RAM · 16 bays · 2 x 10GbE
AINAS Score 50
No verified price

AI CapableQNAP
Why AI Capable: Ryzen V1500B (2018 embedded) is not a current CPU; its PCIe Gen3 x4 slots take only low-profile GPUs up to 6GB on QNAP's list.
Updated
Can the QNAP TS-873A run local AI?
Yes. AMD Ryzen Embedded V1500B, backed by a 64GB memory ceiling, lets the QNAP TS-873A handle 32B class models comfortably, and 70B class models at 4-bit with little headroom (Estimated).
A graphics card can be added through its PCIe x4 slot, which is the biggest single speed upgrade for local models.
Local LLM support is documented by the manufacturer.
The TS-873A suits buyers who want an eight-bay QNAP NAS that can also run containers and small local AI jobs. It has 8 bays, 2 M.2 slots, 8GB of DDR4 SODIMM memory expandable to 64GB, and ECC support. Docker and virtual machines are both supported, and QNAP documents Ollama and Open WebUI through Container Station (Vendor confirmed). Two PCIe Gen3 x4 slots take a low-profile NVIDIA card from the compatibility list. With 64GB fitted, Llama 3.3 70B fits at Q4 (Estimated). It carries 2 x 2.5GbE and a 3 year warranty, and the AINAS Score is 49.
The main catch is the GPU path. The PCIe slots only take low-profile cards, and the vendor list tops out at the RTX 3050 LP 6G, so GPU-run models stay small. There is no NPU and no built-in GPU, so anything not on a card runs on a 4-core Ryzen Embedded V1500B. Buyers who want fast inference on mid-size or large models should look for a system with a full-length GPU slot, a current CPU or NPU, and more video memory. Buyers who need 10GbE should know it only comes through a PCIe add-in card here.
Amazon UK
TS-873A-8G
£1,320
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 | Yes | Vendor confirmed | QNAP Ollama guide (x86, optional NVIDIA GPU from compatibility list). 64 GB RAM; GPUs up to RTX 3050 6GB, so small models on GPU. |
| Ollama | Yes | Vendor confirmed | Container Station Ollama tutorial for x86_64 QTS/QuTS hero. |
| Open WebUI | Yes | Vendor confirmed | QNAP Open WebUI tutorial, QTS 5.2 / QuTS hero h5.2+. |
| RAG | Limited | Unverified | Qsirch AI Knowledge Search (RAG) uses cloud LLMs; not confirmed for this model. |
| AI search | Yes | Vendor confirmed | Qsirch AI Semantic Search needs 64-bit x86 and 8 GB RAM. |
| GPU expansion | Yes | Vendor confirmed | GPGPU list up to RTX 3050 LP 6G; product page: entry-level NVIDIA cards for video processing or VM passthrough. |
| Hardware acceleration | Limited | Vendor confirmed | No iGPU; add-in GPU, QAI-M100/U100 (3 TOPS) and Coral M.2/USB Edge TPU on compatibility list. |
| AI photos | Yes | Vendor confirmed | QuMagie AI-assisted photo management (people, things, places). |
| AI video | Yes | Vendor confirmed | QVR Face listed under QTS. |
| Surveillance AI | Yes | Vendor confirmed | QVR Pro 8 free channels (QTS); QVR Surveillance 2 free + 6 on registration. |
Calculated from the TS-873A'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 TS-873A's use-case scores and capability rows. Unknown means we cannot confirm it, not that it probably works.
The TS-873A is an acceptable local LLM host, not a fast one. QNAP confirms an Ollama guide for it, with up to 64GB of RAM and GPUs up to an RTX 3050 6GB, so GPU-run models stay small.
Private RAG is workable but not fully confirmed. Open WebUI and Ollama are vendor confirmed, while the Qsirch RAG row is limited and Unverified, uses cloud LLMs and is not confirmed for this model.
AI search is an acceptable fit. Qsirch AI Semantic Search is vendor confirmed and needs 64-bit x86 and 8GB of RAM, which the base configuration meets.
AI photo management is an acceptable fit. QuMagie is vendor confirmed for sorting by people, things and places, and it runs on the x86 CPU rather than needing an NPU.
Surveillance is an acceptable fit. QVR Pro offers 8 free channels on QTS, and QVR Surveillance gives 2 free plus 6 on registration, both vendor confirmed. QVR Face is also listed.
Home lab use is acceptable. Docker and virtual machines are both supported, GPU pass-through is listed as working, and there are 8 bays, 2 M.2 slots and ECC memory.
Business use is acceptable. ECC memory, 8 bays, a 3 year warranty and a 250W power supply cover the basics, though network ports are 2 x 2.5GbE unless a card is added.
Creative AI is an acceptable fit with limits. GPU expansion is vendor confirmed, but hardware acceleration is limited: there is no iGPU, and cards are low-profile only, up to the RTX 3050 LP 6G.
Out of the box the TS-873A has 8GB; the official ceiling is 64GB of DDR4 SODIMM with ECC support. A memory upgrade is the cheapest way to make it more useful for AI.
Working backwards from 64GB: 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 93B (Estimated).
ECC support protects against silent memory errors on a box that runs around the clock.
Only a small one. The 2x PCIe Gen3 x4 in the TS-873A is sized for network and storage cards; low-profile GPUs are the limit.
For a personal library, yes. A compact embedding model and a small chat model will run, but the TS-873A is better as the document store behind a RAG system whose model runs on stronger hardware.
The manufacturer also provides AI search across your files (documented by the manufacturer), which covers simple document questions without setting up your own stack.
Built-in networking: 2 x 2.5GbE. Adequate for a household, slower for bulk indexing.
| Configuration | RAM | SSD | Notes |
|---|---|---|---|
| TS-873A-8G | 8GB | None |
The closest QNAP model by bay count and AINAS Score. Only the specifications that differ are listed.
Its strongest AINAS profiles are Home Lab at 53 and RAG at 51, which points to home lab owners who want one box for storage, containers, media and a few AI services.
Look elsewhere if you are teams moving large files over the network every day.
Other AI Capable systems closest in price.
Yes, at Q4 only, and only with the memory upgraded. Llama 3.3 70B needs 43GB at Q4 and fits (Estimated), but it does not fit at Q8, which needs 75GB (Estimated). The system ships with 8GB and takes up to 64GB. Llama 4 Scout 17Bx16E needs 67GB at Q4 and does not fit (Estimated).
Yes. The TS-873A ships with 8GB of DDR4 SODIMM and takes up to 64GB across 2 slots, using 2x 32GB modules. The memory sits in slots rather than being soldered. Fitting 64GB is what lets larger models load, such as Llama 3.3 70B at Q4 (Estimated).
Yes. QNAP lists ECC support for the DDR4 SODIMM slots. The system includes 8GB as standard and supports up to 64GB. ECC support is listed for the memory type, so check module compatibility before buying upgrades.
Yes, but only low-profile cards. It has 2 PCIe Gen3 x4 slots, and QNAP's list includes the GTX 1650 LP, GTX 1630, RTX 3050 LP 6G, Quadro P620 and P1000, and RTX A400 4GB. This is vendor confirmed. Maximum video memory on that list is 6GB.
Yes, it has 2 M.2 2280 slots. They run at PCIe Gen3 x1. They are separate from the 8 SATA bays, and the 2 PCIe Gen3 x4 slots remain free for cards such as a network adapter.
Yes. Container Station runs Docker on QTS 6.1.0 and QuTS hero h6.1.0, and virtual machines are supported as well. QNAP's Ollama tutorial for x86_64 and its Open WebUI tutorial for QTS 5.2 and QuTS hero h5.2 or later are vendor confirmed, so local models run as containers, either on a low-profile NVIDIA card from the compatibility list or on the 4-core Ryzen Embedded V1500B. QNAP also names VM passthrough as a use for entry-level NVIDIA cards. With 8GB included, plan a memory upgrade before running several containers alongside a model.
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, and the 2x PCIe Gen3 x4 can take a faster network card. 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.