Antix
Antix is Antigma's LLM Proxy, Identity Provider, and Organization Manager — the collaborative backend that makes AI scalable, secure, and reliable for teams.
While Ante delivers autonomous AI capabilities to your local terminal, Antix is the control plane: a unified gateway that routes models across multiple wire protocols, governs organizations, issues budget-capped keys, and tracks AI spend across your company.
Key features
Speaks OpenAI Chat Completions, OpenAI Responses, Anthropic Messages, and Gemini native on the same base URL. Point any existing SDK at Antix with one line of config.
Issue scoped sk-antix-… keys with an atomic max_budget cap. Strict enforcement blocks overruns before upstream is called.
Governance for local coding agents. Start ante, type /providers, and sign in to Antix to securely authenticate your local agent, attributing every prompt.
Send your own provider credentials in Authorization and declare the provider with X-Antix-Provider. Antix routes and meters without re-billing.
Why Antix?
- Fail-closed by default. If the billing backend is unreachable, Antix refuses to serve traffic.
- High-performance hot path. A streaming pipeline normalizes SSE across providers, and atomic budget enforcement is guaranteed under concurrent load.
- Multi-tenant from day one. Organizations, RBAC (
admin/member), and scoped virtual keys — not bolted on. - Honest about retention. Antix persists request and response bodies for cost attribution and admin analytics — see Privacy, Security & Data Retention.
Next steps
Make your first request to the Antix proxy in under 5 minutes.
Endpoints, SDK compatibility, and provider overrides.
Provision scoped keys with a hard spend cap.
Manage members, assign roles, and scope access across your organization.
Standardized error codes across providers.