G8 is a self-hosted API gateway and AI gateway: one console to secure, control and observe your API calls, LLM requests and AI agent calls. Built for banks, telecom operators, insurers and public institutions that cannot hand their API and AI traffic to a vendor's cloud. The console, the gateways, the analytics and the audit log all run in your network, on OpenShift, Kubernetes, Docker or virtual machines, including networks without internet access. Nothing calls home: the license is a signed file checked locally. What G8 covers • G8 API Gateway: rate limits, identity with Active Directory, LDAP, SAML or OpenID Connect, and traffic rules for every API. • G8 AI Gateway: one set of rules for traffic to OpenAI, Azure OpenAI, Anthropic, Amazon Bedrock, Google Gemini or any OpenAI-compatible endpoint, with token limits, prompt rules and provider fallback. Existing APIs are served to AI agents as MCP tools from their OpenAPI documents, with the same authentication and limits. • G8 API Security: authentication, request validation and data masking at the gateway, with separate roles for each environment. • G8 Developer Portal: a catalog of your APIs with documentation and self-service access. Change control by design: a change is applied only after its audit record is written, and refused attempts are recorded too. One console manages every gateway cluster, from development to production. Our founders are former enterprise architects who have designed, migrated and run API gateways. Support comes from the engineers who build G8, and help with installation and upgrades is included in the Enterprise license. Evaluate with the free Community edition, or request a demo: g8.dev/request-a-demo Documentation: docs.g8.dev