vARX Series
Private GPU Cloud Platform
A ready-to-use monthly rental service designed for developers, ISVs, and AI SaaS teams who need a private cloud PaaS experience close to a real on-prem AI appliance.
The vARX Series allows teams to quickly access a private GPU compute environment for remote development, POC, model deployment, and performance validation — without the need for a large upfront investment in hardware.
Powered by ARX Edge OS, vARX simplifies GPU server setup, model deployment, API Endpoint creation, API Key management, and ongoing operations. Developers can deploy models and publish service endpoints through a no-code graphical interface, allowing them to focus on application innovation instead of infrastructure complexity.
Once development is complete, applications can be seamlessly migrated to the ARX physical appliance series and deployed to enterprise sites, branch offices, factories, retail locations, and other distributed environments.
Shorten the Deployment Time for On-Prem Models and GenAI Applications
From hours to within 15 minutes!
The biggest challenge in adopting on-prem AI is often not the model itself, but the integration of GPU servers, inference engines, model formats, API endpoints, access control, and operational environments.
With ARX Edge OS, vARX enables developers and IT teams to deploy private models, publish APIs, and test connections through a graphical interface — significantly shortening POC and validation cycles.
Key Features
GUI-based Operation Console
Users can remotely log in through a browser and manage models, endpoints, API keys, applications, GPU resources, and system status through a graphical interface.
One-click Model Deployment
Supports point-and-click import and deployment of open-source models from Hugging Face, with automatic configuration of inference engines and GPU resources.
OpenAI-compatible API Endpoint
Publishes different on-prem AI models as standardized LLM Endpoints for enterprise systems, agents, RAG applications, and developers.
Model Runtime & API Management
Supports on-prem model deployment and service management across different inference runtimes, including GPU resource allocation, API Key creation, revocation, expiration, model-level access control, RPM management, and usage analytics dashboards.
Security Control
Supports allowlists, blocklists, abnormal access blocking, and API access control to help enterprises establish a more secure private AI environment.
Core Positioning
Monthly Rental
Fast Deployment
Remote POC
Private & Secure
Seamless On-Prem Deployment
vARX provides each customer with a dedicated, customer-controlled private AI environment. During the rental period, models, data, and application content remain under the customer’s control. ARDGE does not access, operate, or inspect customer workloads, system activities, or data content.
Storage Option
ARX supports external storage expansion, allowing users to mount large-capacity storage to expand knowledge base file sources and system storage capacity.
Option: NAS with multi-TB storage, mounted over NFS
Edge AI Platform for ISVs and AI SaaS Developers
ARX OS helps developers quickly complete AI application POCs, package applications into containers, and deploy them across multiple on-premises Edge AI devices. Developers focus on application innovation, while ARX OS handles the complexity of deployment and operations.
ARDGE deploys ARX Edge OS on platforms such as NVIDIA GB10 and RTX PRO 6000 GPU systems, providing a ready-to-use environment for on-premises AI development, validation, and deployment.
Software developers no longer need to manage GPU servers, container environments, model configuration, or infrastructure operations. After completing a POC, applications can be packaged and deployed across multiple ARX edge devices, enabling scalable Edge AI solution delivery.
Key Benefits for Developers
- 1
Simplified AI Deployment
Use a graphical, no-code interface to quickly deploy models, create API keys, publish LLM endpoints, and manage AI services.
- 2
From POC to Multi-Site Edge Deployment
Developers can validate and package applications on ARX OS, then deploy containerized applications across multiple Edge AI devices for customers, branch offices, factories, retail locations, and distributed environments.
- 3
GenAI App Store Architecture
ARX OS provides a GenAI App Store framework that allows ISVs to publish applications to the platform. End users can install applications with one click, while developers manage version updates and application lifecycle.
- 4
OTA and Agent Integration
The system supports OTA updates and can integrate with Agent workflows through MCP／CLI, enabling remote management and automated operations.
- 5
Built-in On-Prem AI Toolchain
ARX OS includes tools for on-prem OCR preprocessing, document conversion, and video/audio speech-to-text, helping developers build RAG, knowledge base, and multimodal AI applications faster.
vARX Private GPU Cloud Platform Rental Plans
Private. Secure. Ready to Deploy.
| Plan | Specification — Physical Machine / VRAM or Memory | Ideal For | Monthly Rental |
|---|---|---|---|
 vARX-100 | ARX-100｜GB10-class compact private AI development environment｜Approx. 128 GB Unified Memory | Individual developers, small ISVs, demos, POC development, lightweight RAG, API testing | USD 599 / month |
 vARX-200 | ARX-200｜x86 + RTX PRO 6000-class workstation / server｜Approx. 96 GB–192 GB VRAM | ISV POCs, enterprise validation, multi-model testing, endpoint stress testing | USD 5,999 / month |
 vARX-300 | ARX-300｜High-end AI Station / large-model validation environment｜Approx. 252 GB memory / VRAM-class configuration | Finance, manufacturing, government, large-scale RAG, multi-user and high-performance validation | USD 12,999 / month |
Note: Actual VRAM / memory capacity may vary depending on the delivered system configuration. The above serves as a reference for plan positioning and monthly rental pricing.
Learn more about ARDGE’s latest Private AI solutions:
Download Flyer