Consus
A FedRAMP Equivalent AI gateway that lets US defense contractors use frontier AI models on ITAR and CUI data with zero data retention.
At a Glance
About Consus
Consus, made by Consus Industries, Inc., is an AI gateway for US-based aerospace, defense, and energy contractors that need to use commercial AI models on export-controlled and controlled unclassified data. Access is granted after a company review and contract; there is no self-serve option.
What It Is
Consus is a compliant inference layer that sits between a company's AI tools and upstream model providers. Any tool that can point at a custom API endpoint can use it, and OpenAI-compatible clients can connect to api.consus.io. Requests run on US-only infrastructure with US persons in the data plane, and model IDs carry a suffix such as :itar that names the boundary they run in. The company states that it is FedRAMP Equivalent, assessed against all 410 FedRAMP High controls by a third-party assessor, and that zero data retention is in place with every upstream model provider.
Identity-Driven Model Access
Admins connect Microsoft Entra ID, Okta, Google Workspace, or any OpenID Connect provider and map identity-provider groups to policies. Users get the union of their groups' policies, capped by an org-wide ceiling such as FedRAMP High plus ITAR. Requests outside policy are refused, and policy changes reach every key immediately and are logged.
Rollout and the Portal
The Portal builds deployment packages for Claude Desktop, ChatGPT Desktop, Claude Code, Codex CLI, Grok Build, OpenCode, Pi, and VS Code on macOS and Windows. Admins push these through Jamf, Kandji, Intune, or Group Policy. No API key is placed in any generated file. The open-source Consus Launcher (Apache-2.0) opens five of the tools already configured. The Portal also shows spend and budgets, usage analytics, key ownership, groups and policies, and model controls, and it shows usage metadata but never prompt content.
Management API
A management key cannot run inference, but it can issue and revoke keys, set monthly limits, and read org caps, spend, and request logs (model, tokens, cost, latency, status). This allows integration with onboarding, offboarding, SIEM, and assessor reporting workflows.
Community Discussions
Be the first to start a conversation about Consus
Share your experience with Consus, ask questions, or help others learn from your insights.
Pricing
Teams
Under 50 employees, ITAR workloads. AI tokens billed at cost with no markup. Access granted after company review and contract; no self-serve.
- Works with every AI tool you already use: Claude Code, Codex, ChatGPT, VS Code and more
- ITAR routing for Claude, GPT and Grok models
- ITAR attestation
- Email sign-in with MFA
- Usage dashboard and email support
- + AI tokens at cost, no markup
Enterprise
50+ employees, or any CUI workload. Custom pricing; annual agreement plus tokens at cost.
- Everything in Teams
- CUI routing with CMMC Level 2 support
- Single sign-on (SSO)
- Body of Evidence and SSP inputs for your CMMC assessment
- Usage and audit logs via API, ready for your SIEM
- Dedicated support
- + AI tokens at cost, no markup
Capabilities
Key Features
- Zero data retention with upstream model providers
- ITAR-oriented controls with US-only infrastructure and US persons in the data plane
- FedRAMP Equivalent, assessed at the High baseline
- Identity provider group-to-policy model access
- Org-wide model ceiling by compliance level
- Per-org and per-key monthly budgets
- Portal for spend, usage analytics, keys, and policies
- Ready-made deployment packages for eight tools on macOS and Windows
- Consus Launcher that opens AI tools pre-configured
- Management API for key lifecycle, spend, and request logs
- OpenAI-compatible endpoint
- Metadata-only request logs