OwnAI Compute
Book a call Sign in
Explore AI Stations

Select the station for your organisation

Three specified configurations on Jetson Orin Nano, Orin NX and AGX Orin, plus a custom build to your requirements. Each is a starting point in the configurator; access plans are chosen separately by number of users.

AI Station Mini

A private AI assistant for individuals and small teams.

$1,406 from, excl. tax
Platform
Jetson Orin Nano 8GB
Memory
8 GB 128-bit LPDDR5
AI performance
67 TOPS (INT8, sparse (Super mode))
Storage
1 TB NVMe
Power
10–34 W
  • Sole practitioners
  • Small teams
  • Home offices
  • Students and educators

AI Station Pro

Most popular

Dependable on-premises AI for departments and professional practices.

$3,381 from, excl. tax
Platform
Jetson Orin NX 16GB
Memory
16 GB 128-bit LPDDR5
AI performance
157 TOPS (INT8, sparse (Super mode))
Storage
2 TB NVMe + 2 TB external
Power
15–53 W
  • Professional practices
  • Departments
  • Consultancies
  • Content and media teams

AI Station Ultra

Organisation-grade AI infrastructure for demanding workloads.

$7,876 from, excl. tax
Platform
Jetson AGX Orin 64GB
Memory
64 GB 256-bit LPDDR5
AI performance
275 TOPS (INT8, sparse)
Storage
4 TB NVMe
Power
23–81 W
  • Small and medium businesses
  • Research groups
  • Clinics and legal practices
  • Operations and robotics teams

Custom AI Station

Build your own

Specified to your organisation's requirements.

Pick any Jetson module and only validated parts are offered.

  • Organisations with a defined workload or procurement specification

Side-by-side comparison

Suggested configurations, priced for region US. Final pricing depends on your choices in the configurator.

● AI Station Mini● AI Station Pro● AI Station Ultra● Custom AI Station
Jetson platformJetson Orin Nano 8GBJetson Orin NX 16GBJetson AGX Orin 64GBYour choice
CPU6-core Arm Cortex-A78AE, 1.5 MB L2 + 4 MB L38-core Arm Cortex-A78AE, 2 MB L2 + 4 MB L312-core Arm Cortex-A78AE, 3 MB L2 + 6 MB L3—
GPU1024-core NVIDIA Ampere architecture, 32 Tensor Cores1024-core NVIDIA Ampere architecture, 32 Tensor Cores2048-core NVIDIA Ampere architecture, 64 Tensor Cores—
Peak AI performance67 TOPS (INT8, sparse (Super mode))157 TOPS (INT8, sparse (Super mode))275 TOPS (INT8, sparse)—
Memory8 GB 128-bit LPDDR5, 102 GB/s16 GB 128-bit LPDDR5, 102 GB/s64 GB 256-bit LPDDR5, 204.8 GB/s—
Storage1 TB NVMe2 TB NVMe + 2 TB external4 TB NVMeConfigurable
Enclosure & coolingActive heatsink with PWM fan (Orin Nano / NX), Compact desktop enclosureActive heatsink with PWM fan (Orin Nano / NX), Premium aluminium small-form-factor enclosureAGX Orin active thermal solution, Premium aluminium AGX enclosureConfigurable
Networking1× Gigabit Ethernet; M.2 Key E for Wi-Fi; Wi-Fi 6E + Bluetooth 5.3 module (M.2 Key E)1× Gigabit Ethernet; 2.5 GbE USB-C adapter1× Gigabit Ethernet + 1× 10 Gigabit Ethernet; 10 GbE SFP+ PCIe network card—
Estimated power10–34 W15–53 W23–81 W—
Realistic LLM size7–8B parameter LLMs at INT4, local RAG with embeddings7–8B LLMs at INT8 or 13B at INT4, multiple concurrent services30B-class LLMs at INT4, 70B at INT4 (slow), multi-camera analyticsDepends on module
Included softwarePersonal AI PackagePrivate Knowledge PackageComplete Personal AI PackageYour choice
Best forSole practitioners, Small teams, Home offices, Students and educatorsProfessional practices, Departments, Consultancies, Content and media teamsSmall and medium businesses, Research groups, Clinics and legal practices, Operations and robotics teamsOrganisations with a defined workload or procurement specification
Price (excl. tax)$1,406$3,381$7,876From any budget
ConfigureConfigureConfigureConfigure

Rejoining the server...

Rejoin failed... trying again in seconds.

Failed to rejoin.
Please retry or reload the page.

The session has been paused by the server.

Failed to resume the session.
Please retry or reload the page.