Olares One: computer for local LLMs
Last updated
Olares One is a complete computer for local LLMs with an RTX 5090 Mobile (24 GB of VRAM), an Intel Core Ultra 9 275HX, 96 GB of RAM and 2 TB of storage for 4,999 USD. Olares OS lets it serve models to several devices.
Olares One → Computer for local LLMs → Ollama · vLLM · LM Studio → Local models without the cloud
What hardware does Olares list?
Olares lists a laptop-class NVIDIA GPU, a 24-core Intel CPU and 96 GB of DDR5 in a 320 x 197 x 55 mm box. The GPU is an RTX 5090 Mobile with 24 GB of GDDR7 and a 175 W max TGP. Storage is a 2 TB NVMe SSD. The table uses Olares' figures, read 13 November 2026.
| Part | Specification |
|---|---|
| GPU | NVIDIA GeForce RTX 5090 Mobile, 24 GB GDDR7, 175 W max TGP |
| AI performance | 1824 AI TOPS (Olares' claim) |
| CPU | Intel Core Ultra 9 275HX, 24 cores, up to 5.4 GHz |
| RAM | 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; Wi-Fi 7, Bluetooth 5.4 |
| Power | 330 W adapter |
| Weight | 2.15 kg |
| Price | 4,999 USD list price, Olares shop |
Olares' own benchmarks on this machine show Qwen3-30B-A3B at 157 tokens per second for one request, gpt-oss-120b at about 36 tokens per second with part of the model in system memory, and FLUX.1 dev (Nunchaku FP4) at 8.32 s per image after the first.
Why do we feature it as a computer for local LLMs?
Because most computers bought for local LLMs end up serving more than one device. You chat from a laptop, call the API from a script and check an answer on a phone. Olares One is built for that pattern. Olares OS runs it as a server with accounts, a built-in VPN and one-click installs of Ollama and Open WebUI.
It runs on x86 with NVIDIA CUDA, so vLLM and CUDA-first image and video tools work. According to Olares' FAQ, you can also wipe Olares OS and install Windows or another Linux distribution, which turns it into a standard workstation.
It is the wrong pick for one person who needs 70B and larger models at speed. A Mac Studio or DGX Spark holds more.
How do you start using it?
Activate it, install a model runner and point your tools at it. Activation needs an internet connection, an Olares ID and the LarePass app, according to Olares' FAQ. Ollama and Open WebUI then install from the Olares Market. A 32B model at 4-bit, about 19-20 GB (approximate rule of thumb at 4-bit), is a sensible first download for 24 GB.
Your other tools then call the server's API. LM Studio, for example, runs on your laptop and uses models hosted on Olares One. The Ollama guide covers model choice.
What should you confirm with Olares directly?
Confirm Silent Mode power, desktop use, GPU access in virtual machines and the price. Olares' pages give different GPU power figures for Silent Mode. On Olares OS the HDMI port shows a shell or game output, without a full desktop. Olares' FAQ says GPU passthrough to a Windows VM is not supported, so Windows GPU work needs dual boot.
On price, Olares' shop announced 300 USD off per unit from 17 November 2026, which was not yet live on 13 November.
Questions & answers
Can I use Olares One as a normal PC?
Yes. Olares' FAQ says you can wipe Olares OS and install Windows or any Linux distribution, or dual-boot from a second SSD in the free M.2 slot. It then works as a desktop with an RTX 5090 Mobile.
Can I run LM Studio on Olares One?
Not on the box. Olares calls LM Studio a client-side desktop app, so it lives on your laptop or desktop and reaches the models hosted on Olares One over the API.
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 Computer for Local LLM. “Olares One: computer for local LLMs.” Content reviewed 13 November 2026.
https://bestcomputerforlocalllm.com/olares-one/