Deployment

Your operations. Your infrastructure. Your choice.

ResolveGrid runs wherever your work happens. Fully managed in the cloud, inside your own data center, or completely disconnected from the internet. Same platform, same AI-guided workflows, three ways to deploy.

ResolveGrid PlatformAgentic AI · Computer Vision · Intelligent Workflows · Remote ExpertCloudManaged SaaSChoice of AI providerOn-PremisesYour data center or VPCLocal LLMs on your GPUsAir-GappedFully offline enclaveLocal models · No call-homeSECURE BOUNDARY

Three ways to run ResolveGrid

Every deployment model runs the same platform — agentic AI guidance, computer vision, remote expert support, and intelligent workflows. What changes is where it lives and which models power it.

Cloud

Fastest path to value

Fully managed SaaS on secure cloud infrastructure. Deploy in days, scale elastically, and stay current automatically. Bring your preferred AI provider or select to automatically chose the best models.

  • Managed hosting with regional data residency options
  • Best in class AI providers: Anthropic, OpenAI, Google, Azure
  • Automatic updates, zero maintenance
  • Elastic scale for seasonal field demand

On-Premises

Your data never leaves your walls

The full ResolveGrid platform deployed in your own data center or private cloud. Run local, open-weight LLMs on your own GPUs or connect to approved providers through your own gateway.

  • Runs in your data center or private VPC
  • Local LLM support on your own GPU hardware
  • Integrates with your identity provider and SIEM
  • You approve and schedule every update

Air-Gapped

Full capability. Zero connectivity.

Everything, AI inference, computer vision, knowledge base, workflows, runs entirely inside your isolated network. No internet connection required, ever.

  • 100% offline. Local LLMs and vision models on-site
  • No telemetry, no call-home
  • Signed offline update bundles you verify and install
  • Designed for classified and OT environments

Compare deployment models

 
Cloud
On-Premises
Air-Gapped
Hosting
Managed by ResolveGrid on secure cloud infrastructure
Your data center or private cloud
Your isolated network. Fully disconnected
AI models
Anthropic, OpenAI, Google.
Local open-weight LLMs on your GPUs, or approved external providers
Local LLMs and vision models only, on on-site GPUs
Updates
Automatic and continuous
You approve and schedule
Signed offline bundles, installed on your schedule
Data residency
Regional hosting options
Fully under your control
Never leaves the enclave
Connectivity
Internet required
Internal network; optional outbound
None required, ever
Time to deploy
Days
Weeks
Weeks to months, site-dependent
Ideal for
SMEs, mid-market, and fast-moving enterprise teams
Regulated enterprises with strict data governance
Critical infrastructure, defense, classified and OT environments

Which model fits you?

Cloud

SMEs · Mid-market · Fast-moving enterprise teams

You want results this quarter, not an infrastructure project. Ideal for teams without dedicated platform operations, distributed workforces, and organizations whose compliance needs are met by certified cloud hosting. Most organizations start here.

On-Premises

Regulated enterprises · Healthcare · Financial services · Telecom · Manufacturing

Data governance, residency laws, or internal security policy require your data to stay under your control, while your field teams still get modern AI guidance. Your infrastructure, your rules, full platform capability.

Air-Gapped

Critical infrastructure · Energy & utilities · Defense & government · Isolated OT networks

When the environment can't touch the internet, the AI comes to the environment. Built for sites where disconnection is a requirement, not a preference. Every capability running inside your perimeter.

Frequently asked questions

Do all deployment models have the same features?

Feature parity is a design principle. Workflows, computer vision, and expert escalation work the same everywhere. AI model capability scales with your hardware: cloud deployments can use frontier models, while on-prem and air-gapped deployments run the strongest open-weight models your GPUs support.

Which local LLMs are supported?

We maintain a validated set of open-weight language and vision models sized for different GPU footprints, and update it as the open-model landscape evolves. Our team will match a validated configuration to your hardware during scoping.

Can we start in the cloud and move on-premises later?

Yes. It's the same platform in every model, and your workflows, knowledge base, and configuration migrate with you. Many customers prove value in the cloud, then bring the platform inside for scale-out.

How do air-gapped updates work?

Releases ship as cryptographically signed bundles. Your team transfers them across the boundary using your established process, verifies the signatures, and installs on your schedule. No external connection at any point.

Who operates an on-premises deployment?

Your team, with our runbooks, monitoring guidance, and support. If you'd rather not own day-to-day operations, ask about managed assistance options during scoping.

Not sure which fits?

Bring your security team — we'll walk through the architecture together and map the right deployment model to your environment, your compliance requirements, and your hardware.

Request a demo