96GB max RAM · 5 bays · 10GbE
AINAS Score 65
$919.00
Amazon US · Verified
Ships from the US; delivery and import charges shown at checkout

AI ReadyQNAP
Why AI Ready: 12th Gen Core i9 (i7 and i5 also sold), no NPU, PCIe Gen 4 x16 GPU path limited to low-profile cards up to the 6GB RTX 3050, 64GB memory ceiling, and QNAP does not position it for local AI.
Updated
Can the QNAP TVS-h874 run local AI?
Partly. The QNAP TVS-h874 runs local models on 12th Gen Intel Core i9 (16-core, 8P+8E, 24 threads); Core i7 (12-core, 20 threads) or Core i5-12400 (6-core, 12 threads) on other configurations alone, and its 64GB ceiling allows 32B class models comfortably, and 70B class models at 4-bit with little headroom (Estimated), but replies will be slow without graphics acceleration.
A graphics card can be added through its PCIe x16 slot, which is the biggest single speed upgrade for local models.
Local LLM support is documented by the manufacturer.
The TVS-h874 suits a buyer who wants an 8-bay tower that also hosts local AI in containers. The Core i9 configuration has 16 cores, 64GB of DDR4 SODIMM memory, 2 x 10GbE, 2 x 2.5GbE and two M.2 slots. It runs QTS 6.1.0 or QuTS hero h6.1.0, with Docker and virtual machines supported. Ollama and Open WebUI are vendor confirmed through QNAP Container Station tutorials. Llama 3.3 70B fits at Q4 (Estimated), so the 64GB of memory is the real strength for model size. QuMagie, Qsirch AI Semantic Search and QVR Surveillance are vendor confirmed as well. The 2 x 10GbE ports are on the i9 configuration only. The AINAS Score is 66, and the warranty is 3 years.
The GPU path rules it out for some buyers. The PCIe Gen 4 x16 slot is occupied by the 10GbE card on the i9 as shipped, and the largest listed GPU is the 6GB RTX 3050 LP. Larger models are CPU-bound, and local LLM support is listed as limited. Qsirch AI Mode needs a GPU with 16GB or more of VRAM, and none is on this model's GPU list. Buyers who want large-model work or GPU-backed AI search should look for a system with a free x16 slot that takes a GPU with 16GB or more of VRAM. Check the configuration before buying, as the i7 and i5 versions have no 10GbE.
Amazon UK
TVS-h874-i5-32G
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 | Vendor confirmed | QNAP Ollama guide applies (x86_64, 16GB or more for 7B models, GPU optional and must be in Container Station mode). Largest listed GPU is the 6GB RTX 3050 LP, so larger models are CPU-bound. |
| Ollama | Yes | Vendor confirmed | QNAP Container Station Ollama tutorial for QTS and QuTS hero on x86_64. |
| Open WebUI | Yes | Vendor confirmed | QNAP Open WebUI tutorial, QTS 5.2 or QuTS hero h5.2 or later. |
| RAG | Limited | Unverified | Qsirch lists on-premises RAG with no model named. Not confirmed for this model. |
| AI search | Yes | Vendor confirmed | Qsirch AI Semantic Search: 64-bit x86, 5.0.1 or later, 8GB RAM. AI Mode needs an x74 or newer NAS with a 16GB or larger VRAM GPU; none is on this model's GPU list. |
| GPU expansion | Limited | Vendor confirmed | PCIe Gen 4 x16 slot, 185 mm card limit, 350W PSU. GPU list tops out at RTX 3050 LP 6GB and RTX A400 4GB. TVS-h874X list has RTX 3050 LP BRK 6GB and omits the RTX A400. |
| Hardware acceleration | Yes | Vendor confirmed | UHD Graphics 770 or 730; Coral Edge TPU and Hailo-8 on the AI accelerator list; listed GPUs support Hardware Transcoding. |
| AI photos | Yes | Vendor confirmed | QuMagie listed on QTS and QuTS hero. |
| AI video | Yes | Vendor confirmed | QVR Face listed under QTS only. |
| Surveillance AI | Yes | Vendor confirmed | QVR Surveillance 2 free channels plus 6 on registration; QVR Pro 8 free channels (QTS only). |
Calculated from the TVS-h874'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 TVS-h874's use-case scores and capability rows. Unknown means we cannot confirm it, not that it probably works.
The TVS-h874 is a good fit for local LLM work. Ollama is vendor confirmed through Container Station, local LLM support is listed as limited because larger models are CPU-bound, and Llama 3.3 70B fits at Q4 (Estimated) in 64GB.
Private RAG is an OK fit, with one gap. Open WebUI and Ollama are vendor confirmed, but the Qsirch RAG capability is Unverified: it lists on-premises RAG with no model named and is not confirmed for this model.
AI search is a good fit for semantic search. Qsirch AI Semantic Search is vendor confirmed and needs 8GB of RAM, which the 64GB covers, though AI Mode needs a 16GB VRAM GPU that is not on this model's GPU list.
AI photos are a good fit for the TVS-h874. QuMagie is vendor confirmed on both QTS and QuTS hero, and the 8 drive bays plus 2 M.2 slots give plenty of room to hold a large photo library.
Surveillance is a good fit. QVR Surveillance is vendor confirmed with 2 free channels plus 6 on registration, and QVR Pro gives 8 free channels, but QVR Pro runs on QTS only, as does QVR Face.
Home lab is a good fit. Docker and virtual machines are both supported, and the system has up to 64GB of memory, 8 bays, 2 x 10GbE on the i9 configuration and a second PCIe Gen 4 x4 slot.
Business use is a good fit for the TVS-h874. It has 3 years of warranty, 8 bays, 2 x 2.5GbE on every configuration and a choice of QTS or QuTS hero, though the memory is not ECC.
Creative AI is a good fit, with a GPU limit. The x16 slot takes cards up to the 6GB RTX 3050 LP, hardware acceleration is vendor confirmed, with Coral Edge TPU and Hailo-8 on the accelerator list, and local LLM support is limited.
The TVS-h874 supports up to 64GB of DDR4 SODIMM.
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).
It can. A 2x PCIe Gen 4: slot 1 x16 (10GbE adapter preinstalled on i9), slot 2 x4; cards up to 185 x 131.15 x 18.76 mm is the route, and a card with 16GB or more of its own memory changes what the TVS-h874 can run. Measure the space and check the 350W power budget before ordering.
For a personal library, yes. A compact embedding model and a small chat model will run, but the TVS-h874 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 10GbE + 2 x 2.5GbE. Plenty for a shared knowledge base with several users.
| Configuration | RAM | SSD | Notes |
|---|---|---|---|
| TVS-h874-i7-32G | 32GB | None | Core i7 12C/20T, 32GB, no 10GbE, 350W. |
| TVS-h874-i5-32G | 32GB | None | Core i5-12400 6C/12T, 32GB, UHD Graphics 730, no 10GbE, 250W. |
| TVS-h874X-i9-64G | 64GB | None | Core i9 16C/24T, 64GB, 2x 10GBASE-T card preinstalled, 350W. |
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 70 and Business at 69, which points to home lab owners who want one box for storage, containers, media and a few AI services.
Other AI Ready systems closest in overall score.
96GB max RAM · 5 bays · 10GbE
AINAS Score 65
$919.00
Amazon US · Verified
Ships from the US; delivery and import charges shown at checkout
96GB max RAM · 5 bays · 10GbE
AINAS Score 67
$519.00
Amazon US · Verified
Ships from the US; delivery and import charges shown at checkout
192GB max RAM · 12 bays · 2 x 10GbE
AINAS Score 64
No verified price
Workloads it suits
Yes, at Q4 only (Estimated). Llama 3.3 70B needs 43GB at Q4 and 75GB at Q8, so with 64GB of memory Q8 does not fit (Estimated). This is a calculation, not a benchmark. Local LLM support is listed as limited, and larger models are CPU-bound because the largest listed GPU is the 6GB RTX 3050 LP. Ollama is vendor confirmed through Container Station.
Yes, with limits. It has a PCIe Gen 4 x16 slot with a 185mm card limit and a 350W (i9 and i7) power supply. The largest listed GPUs are the 6GB RTX 3050 LP and the 4GB RTX A400. On the i9 configuration the x16 slot is occupied by the 10GbE card as shipped. Support is vendor confirmed but listed as limited.
Yes, up to 64GB. The memory is DDR4 SODIMM and is not ECC. The i9 configuration ships with 64GB, so it has no room to grow. The i7 and i5 configurations ship with 32GB, which leaves them room up to the 64GB maximum. Memory limits model size, so the 64GB ceiling is why Llama 3.3 70B fits at Q4 but not Q8 (Estimated).
Three. The TVS-h874X-i9-64G has a 16-core Core i9, 64GB, a preinstalled 2x 10GBASE-T card and a 350W supply. The TVS-h874-i7-32G has a 12-core Core i7, 32GB, no 10GbE and 350W. The TVS-h874-i5-32G has a 6-core Core i5-12400 with UHD Graphics 730, 32GB, no 10GbE and 250W. All three take up to 64GB of DDR4 SODIMM, so the Estimated fits on this page are the same once the i7 and i5 are upgraded.
Yes. Docker and virtual machines are both supported, and Ollama and Open WebUI are vendor confirmed through QNAP Container Station tutorials. The system runs QTS 6.1.0 or QuTS hero h6.1.0. Ollama needs x86_64, and QNAP says 16GB or more for 7B models, which the 32GB and 64GB configurations meet. Open WebUI needs QTS 5.2 or QuTS hero h5.2 or later, per QNAP's tutorial.
Two PCIe Gen 4 slots: slot 1 is x16 and slot 2 is x4. Cards must be no larger than 185 x 131.15 x 18.76mm. On the i9 configuration, the 10GbE adapter is preinstalled in slot 1. Listed uses include GPUs, with the 6GB RTX 3050 LP the largest, and AI accelerators such as Coral Edge TPU and Hailo-8. These come from QNAP's compatibility list.
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.
Yes. Hardware acceleration support is documented by the manufacturer. See the capability table above for the source.
It has 2 x 10GbE + 2 x 2.5GbE built in, and the 2x PCIe Gen 4: slot 1 x16 (10GbE adapter preinstalled on i9), slot 2 x4; cards up to 185 x 131.15 x 18.76 mm can take a faster network card. The fastest port is 2 x 10GbE.
Every field records its source. Specifications were last checked on . Spot an error? Send a correction with a manufacturer link.