128GB max RAM · 6 bays · 2 x 10GbE
AINAS Score 75
£774
Amazon UK · Verified

AI Ready45Drives
Why AI Ready: AMD EPYC 8004 or Intel Xeon 4th Gen with no NPU or integrated GPU listed, PCIe Gen5 x16 expansion slots with no documented GPU support, and 1152GB (AMD) or 2048GB (Intel) maximum memory, so it is ready under the Xeon or EPYC rule.
Updated
Can the 45Drives Storinator AV15 run local AI?
Only lightly. Expect 70B class models at higher-quality quantization, or large mixture-of-experts models at 4-bit (Estimated) on the 45Drives Storinator AV15: AMD EPYC 8124P, 8224P, 8324P or 8534P, or Intel Xeon Silver 4410Y, Gold 5418Y or Gold 6430 does the work with up to 2048GB of memory, which is fine for experiments rather than a daily assistant.
Its PCIe x16 slot leaves room to add a GPU when you outgrow CPU inference.
Local LLM support is not yet verified.
The Storinator AV15 suits a buyer who wants a 4U rackmount storage server with a large memory ceiling and room to grow. It has 15 hot-swap bays, up to 2048GB of DDR5 ECC RDIMM memory (Intel configurations), 2 x 10GbE and a choice of EPYC or Xeon processors. Memory is expandable and ECC is supported. It has PCIe x16 slots, 1400W is the highest power supply option, and it runs Rocky Linux 9 or Ubuntu 22+ with Houston UI. Podman containers are available through Houston. It scores 74 on the AINAS Score. On memory size alone, Qwen3 235B A22B fits at Q4 and Q8 (Estimated).
The main catch is the AI path. 45Drives does not document GPU fit, power or inference on this model, and every AI capability row is Unverified. The x16 slots exist, but a card working in this chassis is unconfirmed. There is no NPU and no integrated GPU, so acceleration would need an added card. Buyers who want a known local LLM box should look for documented GPU support, stated card power headroom and a verified AI software path. Buyers who only need storage, virtualisation or a home lab are less affected. The system is built in Base, Enhanced or Turbo configurations, so the memory fitted depends on the build you choose.
No verified price
We have not verified a listing for the 45Drives Storinator AV15 at a retailer we track. The specifications below stand on their own; a price appears here as soon as a listing is confirmed.
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.
Each capability shows how it was confirmed: Verified by AINAS, Vendor confirmed, Community confirmed or Unverified.
| Capability | Supported | Evidence | Note |
|---|---|---|---|
| Local LLM | Limited | Unverified | High core count and memory, but no 45Drives statement on LLM use for this model. |
| Ollama | Limited | Unverified | Containers via Podman on Houston; no Ollama evidence on this model. |
| Open WebUI | Limited | Unverified | Runnable as a container in principle; no evidence found. |
| RAG | Limited | Unverified | No evidence found. |
| AI search | No | Unverified | No AI search feature documented. |
| GPU expansion | Limited | Unverified | PCIe x16 slots present; GPU fit and power not documented. |
| Hardware acceleration | Limited | Unverified | No integrated GPU listed; acceleration would need an added card. |
| AI photos | Limited | Unverified | No bundled photo app; self-hosted containers only, no evidence found. |
| AI video | No | Unverified | No AI video feature documented. |
| Surveillance AI | Limited | Unverified | No bundled surveillance software; self-hosted only, no evidence found. |
Calculated from the Storinator AV15'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 | Q8 | Yes | 98GB | Not measured | Estimated |
| Llama 4 Scout 17Bx16ELlama, 109B | Q8 | Yes | 149GB | Not measured | Estimated |
| Qwen2.5 72BQwen, 72.7B | Q8 | Yes | 101GB | Not measured | Estimated |
| Mistral Large 2 123BMistral, 123B | Q8 | Yes | 164GB | Not measured | Estimated |
One verdict per workload, from the Storinator AV15's use-case scores and capability rows. Unknown means we cannot confirm it, not that it probably works.
We cannot confirm the Storinator AV15 for local LLM use. The capability row is Unverified and limited, and 45Drives makes no statement on LLM use for this model, despite the high core count and memory.
Private RAG on the Storinator AV15 is unconfirmed. The RAG row is Unverified with no evidence found, and Open WebUI is only runnable as a container in principle, so nothing here confirms a working setup.
The Storinator AV15 is a poor fit for AI search. No AI search feature is documented, and the capability row is marked no, though that row is Unverified.
We cannot confirm AI photos on the Storinator AV15. There is no bundled photo app, only self-hosted containers, and the Unverified capability row shows no evidence of one working.
Surveillance is unconfirmed on the Storinator AV15. No surveillance software is bundled, it would be self-hosted only, and the Unverified row records no evidence of it running.
The Storinator AV15 is a good home lab fit. It has 15 bays, ECC memory that can be expanded, 2 x 10GbE and Podman containers through Houston UI.
The Storinator AV15 is a good business fit. It has 15 hot-swap bays, ECC DDR5 memory, a 4U rackmount body with rails, and a 1-year warranty with 3, 5 or 7 years extended.
Creative AI on the Storinator AV15 cannot be confirmed. PCIe x16 slots are present, but GPU fit and power are not documented, and hardware acceleration would need an added card.
The Storinator AV15 can be bought without memory and takes up to 2048GB of DDR5 ECC RDIMM with ECC. Fitting the maximum is what turns it into a serious local AI machine.
Our estimate for the biggest model that fits in 2048GB is about 3400B parameters at 4-bit, after setting aside 8GB for the system. Long context windows reduce that figure (Estimated).
ECC memory support is a plus for always-on systems that also hold important files.
Yes, through a AMD: 3x PCIe Gen5 x16 plus 3x PCIe Gen4 x16 and x8. Intel: 6x PCIe Gen5 x16 and x8. That is the single biggest upgrade for local AI on the Storinator AV15: model layers move into fast graphics memory and replies speed up dramatically. Card size and power are the constraints (1400W supply).
As the storage half of a RAG setup it works well; as the model host it is limited. Small embedding models run fine on the Storinator AV15, while the answering model needs more compute than it has.
For network access it has 2 x 10GbE. Fast enough that the network will not be the bottleneck.
| Configuration | RAM | SSD | Notes |
|---|---|---|---|
| Base, EPYC 8124P or Xeon Silver 4410Y | None | None | |
| Enhanced, EPYC 8224P or 8324P, or Xeon Gold 5418Y | None | None | |
| Turbo, EPYC 8534P or Xeon Gold 6430 | None | None |
The closest 45Drives model by bay count and AINAS Score. Only the specifications that differ are listed.
Best fit: home lab owners who want one box for storage, containers, media and a few AI services. That is where the Storinator AV15 scores highest (Home Lab 81/100), followed by RAG (79/100).
Other AI Ready systems closest in overall score.
Workloads it suits
Yes on memory alone, but this is a calculation, not a benchmark. Qwen3 235B A22B needs 142GB at Q4 and 250GB at Q8, and both fit within the 2048GB maximum (Estimated). The AMD configurations top out at 1152GB. Running it is unconfirmed, because the local LLM row is Unverified.
Possibly, but it is unconfirmed. The system has PCIe x16 slots, with Gen5 x16 slots on both AMD and Intel boards. 45Drives does not document GPU fit or power for this model, and the capability row is Unverified and limited.
Yes. The memory is DDR5 ECC RDIMM and expandable. The maximum is 2048GB on Intel configurations, and the AMD specification table lists 1152GB. The included memory is not published, because it depends on the configuration chosen.
Not Docker itself. Containers are managed through Podman on Houston UI, which the capability notes describe. The system runs Rocky Linux 9 or Ubuntu 22+ with Houston UI. Docker support is recorded as available, but the source says Podman, not Docker.
Yes, one. It is a single M.2 slot at PCIe Gen3 x4 on both boards. Boot drives are mirrored SATA SSDs, so the M.2 slot is separate from the boot setup. The system also has 15 hot-swap bays.
Six, on either platform. The AMD boards have 3x PCIe Gen5 x16 plus 3x PCIe Gen4 x16 and x8 slots, and the Intel boards have 6x PCIe Gen5 x16 and x8 slots. 45Drives does not document GPU fit or card power for any of them on this model, so a graphics card in these slots is Unverified. The highest power supply option is 1400W, and the system also has 2 x 10GbE on board, so a network card is not needed for 10GbE.
Partly. Ollama support is not yet verified. See the capability table above for the source.
Partly. Open WebUI support is not yet verified. See the capability table above for the source.
Partly. AI photos support is not yet verified. See the capability table above for the source.
Partly. Surveillance AI support is not yet verified. See the capability table above for the source.
Partly. Hardware acceleration support is not yet verified. See the capability table above for the source.
It has 2 x 10GbE built in, and the AMD: 3x PCIe Gen5 x16 plus 3x PCIe Gen4 x16 and x8. Intel: 6x PCIe Gen5 x16 and x8 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.