AINAS
QNAP TS-664 6-bay NAS

QNAP TS-664

AI CapableQNAP

Why AI Capable: Intel Celeron N5095 quad-core with UHD Graphics, no NPU, 16 GB DDR4 maximum and only a PCIe Gen3 x2 slot with no GPUs on QNAP lists; Edge TPU and Hailo-8 M.2 accelerators are supported for vision tasks only.

Updated

Can the QNAP TS-664 run local AI?

Partly. The QNAP TS-664 runs local models on Intel Celeron N5095 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 short answer

The TS-664 suits a buyer who wants a six-bay home or small-office NAS first and a light AI add-on second. It holds 6 drives, has 2 M.2 slots, two 2.5GbE ports and 8GB of memory as shipped, expandable to 16GB of DDR4 SODIMM. Docker and virtual machines are supported, and Ollama and Open WebUI are vendor confirmed through Container Station. On AI it is a small-model machine. Qwen3 8B fits at Q4 (Estimated), and Llama 3.2 3B fits at both Q4 and Q8 (Estimated). Qsirch AI semantic search, QuMagie photo albums and QVR surveillance are vendor confirmed. It carries a 3 year warranty and an AINAS Score of 34, and it runs QTS 6.1.0 or QuTS hero h6.1.0.

The memory ceiling and the lack of a GPU path rule the TS-664 out for anyone planning to run larger models. Memory tops out at 16GB and the only expansion slot is PCIe Gen3 x2 with no GPUs on the vendor compatibility list, so inference is CPU only. Gemma 3 12B at Q4 is marginal (Estimated), and Mistral Small 3 24B does not fit at Q4 (Estimated), nor do the 32B models, which need 20GB at Q4. Buyers who want those models should look for a system with a GPU expansion path or an NPU, and memory well above 20GB. Buyers who want ECC memory should also note that the TS-664 does not support it, whichever module they fit.

Current offers

  • Amazon US

    TS-664-8G

    Ships from the US; delivery and import charges shown at checkout

    $799.00

    Verified

    In stock

    Buy on Amazon

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.

Price history

Current
$799.00
30-day average
$799.00
90-day low
$799.00
All-time tracked low
$799.00
$794.00$799.00$804.0020 AugToday
Last in-stock Amazon US price we verified each day. Dashed line: 30-day average. Days out of stock are not drawn as prices.

Key specifications

CPU
Intel Celeron N5095
Source: Manufacturer
GPU
Intel UHD Graphics
Source: Manufacturer
Included RAM
8GB DDR4 SODIMM
Source: Manufacturer
Maximum RAM
16GB
Source: Manufacturer
Networking
2 x 2.5GbE
Source: Manufacturer
PCIe
1x PCIe Gen3 x2
Source: Manufacturer
GPU expansion
None
Source: Official docs
Power supply
120W
Source: Manufacturer
OS
QTS 6.1.0 / QuTS hero h6.1.0
Source: Manufacturer
Docker
Yes
VM support
Yes
Released
2022
Segment
Desktop

Can the TS-664 run local LLMs?

Each capability shows how it was confirmed: Verified by AINAS, Vendor confirmed, Community confirmed or Unverified.

CapabilitySupportedEvidenceNote
Local LLMLimitedVendor confirmedQNAP Ollama guide (x86): 8 GB for 3B, 16 GB for 7B models; 16 GB max, CPU only.
Ollama YesVendor confirmedContainer Station Ollama tutorial for x86_64 QTS/QuTS hero.
Open WebUI YesVendor confirmedQNAP Open WebUI tutorial, QTS 5.2 / QuTS hero h5.2+.
RAGLimitedUnverifiedQsirch AI Knowledge Search (RAG) uses cloud LLMs; not confirmed for this model.
AI search YesVendor confirmedQsirch AI Semantic Search needs 64-bit x86 and 8 GB RAM (8G model qualifies; 4G needs upgrade).
GPU expansion NoVendor confirmedPCIe Gen3 x2 slot only; no GPUs on compatibility list.
Hardware acceleration YesVendor confirmedIntel UHD Graphics video conversion; Coral Edge TPU and Hailo-8 M.2 (26 TOPS) on AI accelerator list.
AI photos YesVendor confirmedQuMagie albums by people, objects, locations; Edge TPU speeds recognition.
AI video YesVendor confirmedQVR Face real-time facial recognition on product page.
Surveillance AI YesVendor confirmedQVR Pro 8 free channels (QTS); QVR Surveillance 2 free + 6 on registration.

What AI models can the TS-664 run?

Calculated from the TS-664's memory and each model's published size at each quantization. Rows marked Estimated are calculations, not benchmarks. Check any model.

ModelBest quantizationRuns?Memory neededExpected performanceVerification
Llama 3.2 3BLlama, 3.21BQ8Yes13GBNot measuredEstimated
Llama 3.1 8BLlama, 8.03BQ5Yes15GBNot measuredEstimated
Qwen3 8BQwen, 8.19BQ4Yes15GBNot measuredEstimated
Gemma 3 12BGemma, 11.8BQ4Marginal15GBNot measuredEstimated
Qwen3 14BQwen, 14.8BQ4Marginal16GBNot measuredEstimated
Phi-4 14BPhi, 14.7BQ4Marginal16GBNot measuredEstimated
Mistral Small 3 24BMistral, 23.6BNone fitsNo25GB at Q4Not measuredEstimated
Gemma 3 27BGemma, 27BNone fitsNo29GB at Q4Not measuredEstimated
Qwen3 30B A3BQwen, 30.5BNone fitsNo31GB at Q4Not measuredEstimated
Qwen3 32BQwen, 32.8BNone fitsNo32GB at Q4Not measuredEstimated
DeepSeek-R1 32B (Qwen distill)DeepSeek, 32.8BNone fitsNo32GB at Q4Not measuredEstimated
Llama 3.3 70BLlama, 70.6BNone fitsNo60GB at Q4Not measuredEstimated
Llama 4 Scout 17Bx16ELlama, 109BNone fitsNo89GB at Q4Not measuredEstimated
Qwen2.5 72BQwen, 72.7BNone fitsNo65GB at Q4Not measuredEstimated
Mistral Large 2 123BMistral, 123BNone fitsNo96GB at Q4Not measuredEstimated

Workload fit

One verdict per workload, from the TS-664's use-case scores and capability rows. Unknown means we cannot confirm it, not that it probably works.

Local LLMPoor fit

Poor for local LLM work, and support is only limited. QNAP's Ollama guide is CPU only with a 16GB maximum, giving 8GB for 3B and 16GB for 7B models, and Qwen3 8B fits at Q4 (Estimated).

Private RAGPoor fit

Poor for private RAG, and a fully local setup is not confirmed. Ollama and Open WebUI are vendor confirmed, but Qsirch AI Knowledge Search uses cloud LLMs and is unconfirmed for the TS-664.

AI searchPoor fit

Poor in this rating, even though the feature itself is confirmed. Qsirch AI Semantic Search is vendor confirmed and needs 64-bit x86 and 8GB RAM, which the 8GB TS-664 meets. The 4G configuration would need an upgrade.

AI photosPoor fit

Poor overall, though the photo software is confirmed. QuMagie sorts albums by people, objects and locations (vendor confirmed), and a Coral Edge TPU is listed to speed recognition, but the Celeron has no NPU of its own.

SurveillancePoor fit

Poor in this rating, though camera software is available. QVR Pro gives 8 free channels on QTS, and QVR Surveillance gives 2 free plus 6 on registration, both vendor confirmed, but there is no NPU or GPU path.

Home labPoor fit

Poor on this AI-focused rating. Docker and virtual machines are both supported, and there are 6 bays and 2 M.2 slots, but the 16GB memory ceiling limits how many workloads can run together.

BusinessPoor fit

Poor for AI-led business use. The TS-664 has no ECC memory, memory capped at 16GB and two 2.5GbE ports as its fastest networking, though it does carry a 3 year warranty.

Creative AIPoor fit

Poor for creative AI work. There is no GPU expansion, as the PCIe Gen3 x2 slot has no GPUs on the vendor list, and acceleration stops at Intel UHD video conversion and M.2 accelerators such as Hailo-8 at 26TOPS.

How much RAM does the TS-664 support?

Out of the box the TS-664 has 8GB; the official ceiling is 16GB of DDR4 SODIMM. 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).

Can you add a GPU to the TS-664?

There is no supported way to add a GPU to the TS-664: no PCIe slot and no OCuLink. Inference runs on the CPU, with Intel UHD Graphics handling media tasks, so keep to small models.

Is the TS-664 suitable for RAG?

For a personal library, yes. A compact embedding model and a small chat model will run, but the TS-664 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.

Limitations

  • Inference is CPU only. The PCIe Gen3 x2 slot has no GPUs on QNAP's compatibility list, and there is no NPU, so larger models have no acceleration path.
  • Memory is capped at 16GB. Gemma 3 12B is only marginal at Q4 (Estimated), and models from Mistral Small 3 24B upward do not fit at Q4 (Estimated).
  • The TS-664 has no ECC memory support, and its fastest built-in networking is two 2.5GbE ports. 5GbE or 10GbE needs a card in the single PCIe slot.
  • Qsirch AI Knowledge Search is Unverified for the TS-664 and uses cloud LLMs, so a local RAG workflow is unconfirmed.
  • The M.2 slots run at PCIe Gen3 x1, and there are no USB4, Thunderbolt or OCuLink ports, so fast external storage or an external GPU cannot be added.

Which TS-664 configurations are sold?

ConfigurationRAMSSDNotes
TS-664-4G4GBNone
TS-664-8G8GBNone

How is the TS-664 different from the TS-h686?

The closest QNAP model by bay count and AINAS Score. Only the specifications that differ are listed.

  • Processor: Intel Celeron N5095 on the TS-664, Intel Xeon D-1602 (2-core/4-thread, up to 3.2 GHz) on the TS-h686
  • CPU cores: 4 on the TS-664, 2 on the TS-h686
  • Integrated graphics: Intel UHD Graphics on the TS-664, none on the TS-h686
  • Memory ceiling: 16GB on the TS-664, 128GB on the TS-h686
  • ECC: no on the TS-664, yes on the TS-h686
  • Networking: 2 x 2.5GbE on the TS-664, 4 x 2.5GbE on the TS-h686
  • GPU path: None on the TS-664, PCIe x8 slot on the TS-h686
  • Power supply: 120W on the TS-664, 250W on the TS-h686
  • Status: tracked on the TS-664, discontinued on the TS-h686
  • AINAS Score: 34 on the TS-664, 53 on the TS-h686

See the full QNAP TS-h686 page

Who should buy the TS-664?

Its strongest AINAS profiles are Home Lab at 39 and Business at 38, 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.

What are the alternatives to the TS-664?

Other AI Capable systems closest in price.

Synology RS2821RP+

Synology RS2821RP+

Synology

AI CapableDiscontinued

32GB max RAM · 16 bays · 1GbE

AINAS Score 34

No verified price

View
Synology DS1825+

Synology DS1825+

Synology

AI Capable

32GB max RAM · 8 bays · 2 x 2.5GbE

AINAS Score 35

No verified price

View
TerraMaster F2-424

TerraMaster F2-424

TerraMaster

AI Capable

32GB max RAM · 2 bays · 2 x 2.5GbE

AINAS Score 33

Price awaiting re-verification

Last checked

View

Common questions

Can the TS-664 run Qwen3 8B?

Yes, at Q4 (Estimated). Qwen3 8B needs 5.2GB at Q4 and fits in the 8GB installed (Estimated). At Q8 it needs 8.9GB and is marginal (Estimated), so it loads with little headroom. Inference is CPU only, as QNAP's Ollama guide describes. Larger models such as Gemma 3 12B are marginal at Q4 (Estimated), so Qwen3 8B is the largest model listed on this page that fits at Q4.

Can you upgrade the memory in the TS-664?

Yes. The TS-664 has two DDR4 SODIMM slots, ships with 8GB and takes up to 16GB (2x 8GB). The memory sits in slots rather than being soldered. It does not support ECC. Even at 16GB, Qwen3 32B needs 20GB at Q4 and does not fit (Estimated), so the upgrade helps with 7B to 8B models and Q8 headroom rather than larger ones.

Can the TS-664 take a graphics card?

No. QNAP lists no GPUs for the TS-664, and its only slot is PCIe Gen3 x2. There is no OCuLink, Thunderbolt or USB4 port either. The supported accelerators are Coral Edge TPU and Hailo-8 M.2 (26 TOPS), which are listed for vision tasks only, not for running language models.

Does the TS-664 run Docker?

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 both vendor confirmed, so local language models run as containers. Inference stays on the 4-core Celeron N5095, because the PCIe Gen3 x2 slot takes no GPU. Qsirch AI Semantic Search also needs 8GB of memory, which the TS-664-8G meets.

Does the TS-664 have M.2 slots?

Yes, two. They are M.2 2280 PCIe Gen3 x1 slots, alongside six 3.5in SATA bays. QNAP also lists Coral Edge TPU and Hailo-8 M.2 accelerators as supported, so the slots can hold an accelerator for vision tasks instead of an SSD. Those accelerators do not run language models.

Can the TS-664 run Ollama?

Yes. Ollama support is documented by the manufacturer. See the capability table above for the source.

Does Open WebUI work on the TS-664?

Yes. Open WebUI support is documented by the manufacturer. See the capability table above for the source.

Does the TS-664 do AI photo recognition?

Yes. AI photos support is documented by the manufacturer. See the capability table above for the source.

Can the TS-664 run AI surveillance?

Yes. Surveillance AI support is documented by the manufacturer. See the capability table above for the source.

Does the TS-664 have hardware AI acceleration?

Yes. Hardware acceleration support is documented by the manufacturer. See the capability table above for the source.

How fast is the TS-664's network?

It has 2 x 2.5GbE built in, and the 1x PCIe Gen3 x2 can take a faster network card. The fastest port is 2 x 2.5GbE.

Where do these specifications come from?

Every field records its source. Specifications were last checked on . Spot an error? Send a correction with a manufacturer link.