16GB max RAM · 4 bays · 2 x 2.5GbE
AINAS Score 32
£636
Amazon UK · Verified

AI CapableTerraMaster
Why AI Capable: Intel Core 3 N350 or Core i3-N305 (8 cores, Intel Graphics, no NPU) with 32GB max and no PCIe, USB4 or OCuLink GPU path; TerraMaster markets it as an AI NAS but its OpenClaw agent uses cloud model APIs, so local inference is CPU/iGPU-only via Docker.
Updated
Can the TerraMaster F4-425 Pro run local AI?
Partly. The TerraMaster F4-425 Pro runs local models on Intel Core 3 N350 / Intel Core i3-N305 alone, and its 32GB ceiling allows models up to the 14B class, with 32B class models a tight fit (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 F4-425 Pro suits a buyer who wants a four-bay home or small-office NAS with fast networking and light, CPU-only AI tasks. It has 4 bays, 3 M.2 slots, 2 x 5GbE and up to 32GB of DDR5 memory, with 16GB included on the N350 version. TOS 7 runs Docker and virtual machines. Ollama and Open WebUI are vendor confirmed as Docker containers. TOS 7 Photos adds AI face and scene recognition, and Surveillance Manager handles up to 32 ONVIF cameras. The AINAS Score is 37, which reflects a general-purpose NAS with modest AI ability. Qwen3 32B fits at Q4 (Estimated), though only on the CPU and integrated graphics.
The main limit is the lack of any GPU path. The F4-425 Pro has no PCIe slot, no USB4, no Thunderbolt and no OCuLink, so you cannot add a graphics card later. Anyone who wants local LLMs, private document search or image generation at usable quality is affected. TerraMaster's OpenClaw agent also relies on a cloud provider's API key rather than a local model. Look instead for a system with a PCIe x16 slot or an OCuLink or Thunderbolt port for an external GPU, plus an NPU if you want on-device acceleration. Memory above 32GB would also widen your model choice.
Amazon UK
N305 8GB
£639.99
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/iGPU-only inference via Docker; no NPU or GPU expansion. TerraMaster blog describes Ollama/Open WebUI via Docker on x86 TOS NAS. HWBusters review shows the built-in OpenClaw agent needs an external provider (tested with OpenAI gpt-4o-mini). |
| Ollama | Yes | Vendor confirmed | Runs as a Docker container via TOS Docker Manager per TerraMaster blog (x86 TOS NAS generally). |
| Open WebUI | Yes | Vendor confirmed | Open WebUI container via Docker; same TerraMaster blog source. |
| RAG | Limited | Unverified | Possible through Open WebUI or similar containers; no TerraMaster or community source showing RAG on TOS. |
| AI search | Limited | Vendor confirmed | TOS 7 OpenClaw agent and AI tagging allow natural-language queries; setup uses an API key from a provider such as OpenAI, so cloud LLM. Global search is an inverted index, not semantic. |
| GPU expansion | No | Vendor confirmed | Spec lists no PCIe slot; no USB4, Thunderbolt or OCuLink. |
| Hardware acceleration | Yes | Vendor confirmed | Hardware transcoding engine H.264/H.265/MPEG-4/VC-1 up to 4K60 via Intel iGPU. Video only, no NPU. |
| AI photos | Yes | Vendor confirmed | TOS 7 Photos: AI face recognition and AI scene recognition (80+ categories). |
| AI video | No | Unverified | No AI video feature found in TOS 7 material. |
| Surveillance AI | Yes | Vendor confirmed | TOS 7 Surveillance Manager supports up to 32 ONVIF IP cameras. |
Calculated from the F4-425 Pro'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 F4-425 Pro's use-case scores and capability rows. Unknown means we cannot confirm it, not that it probably works.
Local LLM use is a poor fit. Support is limited and vendor confirmed, with CPU and integrated graphics inference only via Docker, no NPU and no GPU expansion.
Private RAG is a poor fit. The RAG capability is limited and Unverified, so it is unconfirmed, and nothing from TerraMaster or the community shows it working on TOS.
AI search is a poor fit. The OpenClaw agent needs an API key from a cloud provider, and the built-in global search is an inverted index, not semantic.
AI photos is an acceptable fit. TOS 7 Photos offers AI face recognition and AI scene recognition across 80+ categories, which TerraMaster confirms.
Surveillance is an acceptable fit. TOS 7 Surveillance Manager supports up to 32 ONVIF IP cameras, vendor confirmed, and the Intel iGPU handles video transcoding.
Home lab use is an acceptable fit. Docker and virtual machines are both supported, and 3 M.2 slots with 2 x 5GbE give room for containers and fast storage.
Business use is an acceptable fit. It has 4 bays, a 2-year warranty and 2 x 5GbE, but memory is non-ECC and tops out at 32GB.
Creative AI is a poor fit. GPU expansion is confirmed as not supported, with no PCIe slot, USB4, Thunderbolt or OCuLink, and the video acceleration covers transcoding only.
Out of the box the F4-425 Pro has 16GB; the official ceiling is 32GB of DDR5 non-ECC SODIMM (N350); N305 type not stated. A memory upgrade is the cheapest way to make it more useful for AI.
Working backwards from 32GB: 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 40B (Estimated).
There is no supported way to add a GPU to the F4-425 Pro: no PCIe slot and no OCuLink. Inference runs on the CPU, with Intel Graphics (N350) / Intel UHD Graphics (N305) 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 F4-425 Pro is better as the document store behind a RAG system whose model runs on stronger hardware.
Built-in networking: 2 x 5GbE. Adequate for a household, slower for bulk indexing.
| Configuration | RAM | SSD | Notes |
|---|---|---|---|
| N305 8GB | 8GB | None | 1x 8GB, 1 slot. |
| N350 16GB | 16GB | None | 1x 16GB DDR5, 1 slot. |
The closest TerraMaster model by bay count and AINAS Score. Only the specifications that differ are listed.
Its strongest AINAS profiles are Home Lab at 41 and Business at 41, which points to home lab owners who want one box for storage, containers, media and a few AI services.
Look elsewhere if you are buyers who expect to add a graphics card later, or teams moving large files over the network every day.
Other AI Capable systems closest in price.
Yes, at Q4 only. DeepSeek-R1 32B (Qwen distill) is the largest model listed on this page that fits at Q4 (Estimated). It needs 20GB at Q4 and 35GB at Q8, so Q8 does not fit (Estimated). Inference would run on the CPU and integrated graphics through Docker, as there is no NPU or GPU path. Llama 3.3 70B does not fit at either level (Estimated).
Yes. Memory is expandable up to 32GB, in a single SODIMM slot. The N350 version comes with 16GB of DDR5 and the N305 version with 8GB. The N305 memory type is not stated. The memory is non-ECC, so there is no error correction at any size.
No. TerraMaster lists no PCIe slot, and there is no USB4, Thunderbolt or OCuLink port either. That rules out internal and external GPUs. The Intel integrated graphics handle video transcoding up to 4K60 in H.264 and H.265, but this is video only and not general AI acceleration.
Yes, it has 3 M.2 NVMe slots alongside its 4 drive bays. That allows SSD storage or caching without using a bay. The network side is 2 x 5GbE, which suits fast SSD transfers. No SSD is included in either listed configuration.
Two. The N305 8GB pairs an Intel Core i3-N305 with 8GB in a single slot, and the N350 16GB pairs an Intel Core 3 N350 with 16GB of DDR5 non-ECC SODIMM, also in a single slot. Each comes in Gunmetal Grey or Silver. Both processors have 8 cores, and the 32GB ceiling on this page is the system figure. The memory type for the N305 version is not stated by TerraMaster.
Yes. TOS 7 includes Docker Manager, and TerraMaster confirms Ollama and Open WebUI run as containers on it. Inference uses the CPU and integrated graphics only. Virtual machines are also supported. The OpenClaw agent needs an external provider API key, so it does not run a model locally.
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 5GbE built in. The fastest port is 5GbE.
Every field records its source. Specifications were last checked on . Spot an error? Send a correction with a manufacturer link.