Unified AI
Inference Proxy
Centralized authentication, audit logging, and intelligent token budgets. The ultimate gateway for enterprise intranet use, managing multiple models and vendors securely.
Built for the Enterprise
Advanced routing and safety controls built into every request.
Central Auth & Audit
One downstream gateway API key per user. Comprehensive SQLite-backed audit trails for full observability across your organization.
Intelligent Token Budgets
Set central token budgets per user and month. Accurate usage events calculate accounting from upstream, estimation, or failures.
Tenant-level Content Filters
Built-in integration for input review models. Enforce or monitor policies before sending data to upstream LLM vendors.
Smart Routing & Residency
Map public models to any upstream vendor. Maintain compliance with data residency controls enforced via tenant policies.
The Killer Feature: Built-in Management Console
Token Gateway ships with a lightning-fast, server-rendered dashboard built with Hono JSX. No external dependencies, no complex setup. Just powerful control out of the box.
- ✔ Real-time usage tracking & charts
- ✔ Self-service API key management
- ✔ Admin control over budgets & tenants
Proactive AI Guardrails
Stop sensitive data leaks and enforce compliance before requests ever reach upstream vendors. Token Gateway seamlessly integrates with local content review models to provide bulletproof security layers.
- ✔ Enforce or Monitor-only modes
- ✔ Tenant-specific filtering policies
- ✔ Built-in local model integration
Drop-in Replacement
Token Gateway supports standard OpenAI routes including /v1/chat/completions, complete with stream: true SSE proxying.
- ✔ No SDK changes required
- ✔ Supports Embeddings & Completions
- ✔ Built-in Rate Limiting