TokenHub is an open-source enterprise AI gateway published on GitHub that gives companies one private, governed entry point for every model call. The project has gathered around 1,400 GitHub stars and is actively developed, with its latest updates in 2026.
As employees adopt ChatGPT, Claude, coding agents and cheap API relays, model access fragments and cost, data and audit control break down. TokenHub pulls everything back behind a single gateway: it connects OpenAI, Azure OpenAI, Anthropic, Gemini, DeepSeek, Qwen, local models and custom OpenAI-compatible upstreams through native provider adapters and 150+ provider templates, then applies model routing, failover, project-scoped API keys, quotas and concurrency controls on top. Administrators manage providers, routing policies, SSO/OIDC identity, RBAC and audit trails from a web console, while finance and platform teams get usage analytics and provider-bill reconciliation attributed by user, team, project, model and cost center. It deploys privately with Docker Compose or a native systemd install, starts SQLite-first and scales to PostgreSQL for multi-instance setups.
TokenHub is free and open source under the Apache 2.0 license. You self-host it on your own infrastructure and only pay your model providers.
Platform and IT teams at companies rolling out AI tools at scale who need one auditable, cost-controlled gateway instead of scattered personal API keys.