> ## Documentation Index
> Fetch the complete documentation index at: https://docs.altrum.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# AI Gateway

> Send your applications' and coding agents' model traffic through Altrum, so every request is governed.

The AI Gateway is one address for every model call your organization makes. Your applications, SDKs and coding agents point at Altrum instead of at the model provider, and keep working as they did. Altrum passes each request on to your provider, and on the way it:

* checks the request and the reply against your [guardrails](/guardrails/at-runtime.md);
* checks the key, the models it may use, and its [budgets and rate limits](/ai-gateway/budgets-and-limits.md);
* sends it to the right provider, with [fallbacks](/ai-gateway/models-and-routing.md) if that provider fails;
* adds your [MCP tools](/ai-gateway/mcp.md), [saved prompts](/ai-gateway/prompts.md) and [cache](/ai-gateway/semantic-cache.md);
* records what happened, and what it cost, in the [request log](/ai-gateway/request-logs.md).

> **[Send your first request](/ai-gateway/quickstart.md)**
>
> A provider credential, a key and one request, in five minutes.

> **[API reference](/api-reference/introduction.md)**
>
> Every endpoint, header and error.

## Start before you register anything

You can use the AI Gateway as soon as your organization exists. Add a provider credential, create a key, and send requests; your [organization guardrails](/guardrails/organization.md) govern that traffic. When you later register the AI system it belongs to and it has been approved and gone live, attach the key, and that AI system's own guardrails take over. Attaching before it is live stops the key's traffic until it is: see [Virtual keys](/ai-gateway/virtual-keys.md#attach-an-organization-key-to-an-ai-system).

## What it supports

* **Request formats**: OpenAI chat completions and Responses, Anthropic Messages, and Gemini. See the [API reference](/api-reference/introduction.md).
* **Model providers**: OpenAI, Anthropic, Bedrock, Google Gemini, Google Vertex AI and Azure OpenAI, each with your own credentials. See [Provider credentials](/ai-gateway/provider-credentials.md).
* **Clients**: the OpenAI, Anthropic and Google SDKs, Claude Code, Codex CLI, Gemini CLI, Cursor, and any MCP client. See [SDKs](/ai-gateway/sdks.md) and [Coding agents](/ai-gateway/coding-agents.md).
* **Streaming**, in every format.

## Where you manage it

* **AI Gateway** in the sidebar holds keys, provider credentials, the organization's guardrails, MCP servers, prompts, the request log and settings. Its **Overview** has the address and code to copy, and the gateway's [health](/ai-gateway/request-logs.md#health).
* Each AI system's **AI Gateway** tab holds that AI system's keys, limits, MCP tools, cache and content logging, and its requests.
* **Reporting** shows what the traffic costs and what the guardrails caught. See [Reporting](/reports.md).

## Who can do what

Owners and admins run the AI Gateway. Governance Leads can see everything in it, including the request log. Members see **My keys** and **My connections**: when an admin allows personal keys, a member creates one for their own coding agent and connects their own accounts to the organization's MCP servers. See [Virtual keys](/ai-gateway/virtual-keys.md).


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.