Olares One: local AI server
Last updated
Olares One is a local AI server from Olares with an RTX 5090 Mobile, 24 GB of VRAM, 96 GB of RAM and Olares OS preinstalled, listed at 4,999 USD. We feature it as the ready-made choice for a home or small office.
Olares One → Local AI server → Ollama · vLLM · Open WebUI → Private AI on your own network
What does Olares publish about Olares One?
Olares publishes a spec sheet, a FAQ and benchmark results on its own blog. The table takes its numbers from Olares' spec page, product page and FAQ, read on 13 Nov 2026. Noise and AI TOPS figures are Olares' own measurements and ratings, and no independent test of them is in our data.
| Part | Olares One |
|---|---|
| GPU | NVIDIA GeForce RTX 5090 Mobile, 24 GB GDDR7, max TGP 175 W |
| CPU | Intel Core Ultra 9 275HX, 24 cores, up to 5.4 GHz |
| Memory | 96 GB DDR5-5600 (2 x 48 GB), upgradable to 128 GB |
| Storage | 2 TB NVMe SSD, second M.2 slot free |
| Ports | Thunderbolt 5 (80 Gbps), 2.5 GbE, HDMI 2.1, USB-A 3.2 Gen 2 |
| Wireless | Wi-Fi 7, Bluetooth 5.4 |
| Power | 330 W adapter, 30 W in standby |
| Size and weight | 320 x 197 x 55 mm, 2.15 kg (3.1 kg with power supply) |
| Software | Olares OS, open source, built on Kubernetes |
| Price | 4,999 USD list (Olares shop, 13 Nov 2026) |
Olares rates the GPU at 1824 AI TOPS. In its lab test, the box measured 19 dB at idle and stayed under 39 dB at full load.
Why is it featured as a local AI server?
It is featured because Olares OS makes it a shared server on the first day. According to Olares' FAQ, each person gets an account, apps install from the Olares Market in one click, remote access runs through a built-in VPN or reverse proxy, and backups run automatically and encrypted.
Account roles cover Super Admin, Admin and Member, and people connect through the LarePass app. In Olares' own test, Qwen3-30B-A3B ran at 157 tokens/s for one request and about 81 tokens/s per request with 8 concurrent requests. Olares says the box can run offline on the local network after a one-time activation.
How do you get started?
Connect it by Ethernet, activate it once online with an Olares ID and the LarePass app, and create an account for each person. Then install Ollama and Open WebUI from the Olares Market and pull a model that fits in 24 GB of VRAM before anyone starts using it.
The GPU guide lists which model sizes fit. The remote access guide covers phones and laptops away from the office.
What should you confirm with Olares directly?
Confirm the current price, shipping to your country, the apps you need, disk encryption and storage slot details. Olares' own pages disagree on which M.2 slot the stock SSD uses, and according to Olares' FAQ, Olares OS does not encrypt the full disk by default.
- On 13 Nov 2026 the shop banner announced a saving of 300 USD per unit from 17 Nov 2026.
- Olares lists free shipping to the US, EU, UK, Canada, Japan, South Korea and Hong Kong.
- Olares says full-disk encryption is planned as a user option.
- If you need a tool beyond Ollama and Open WebUI, check that the Olares Market lists it.
Questions & answers
Who makes Olares One?
Olares, the company behind the open-source Olares OS. The company named in the shop footer is Bytefuture (HK) Limited.
Is Olares OS free?
According to Olares' FAQ, Olares OS and system updates are free, with no extra fees to use the box. You pay once for the hardware.
Sources for this entry
- Olares One product page (Company-published)
- Olares One specifications (Olares docs) (Company-published)
- Olares One FAQ (Olares docs) (Company-published)
- Olares blog: local AI hardware benchmarking (Oct 2025) (Company-published)
- Olares on GitHub (read via GitHub API 13 Nov 2026) (Company-published)
Continue this reading path
How the related websites connect
Cite this entry
Best Local AI Server. “Olares One: local AI server.” Content reviewed 13 November 2026.
https://bestlocalaiserver.com/olares-one/