[◂ Back to docs](https://zato.io/docs/index.html)

✦

# LLM connections

An LLM connection lets your Python services call Claude, OpenAI, Gemini or any self-hosted model behind one API. The provider's credentials stay in the platform, every call leaves an audit trail, and chat history, skills and GitOps come built in.

[Make your first call](https://zato.io/docs/ai/llm/invoke.html)

## Start here

[### AI gateway overview The architecture-level view of LLM and MCP traffic in one platform](https://zato.io/ai-gateway.html) [### Creating connections Define a connection in the Dashboard - provider, model, address, API key and limits](https://zato.io/docs/ai/llm/connections.html) [### Invoking models The one-shot invoke call from a Python service, and what the response contains](https://zato.io/docs/ai/llm/invoke.html)

## Calling models

[### Invoke - one-shot calls Call a model from a service with self.llm, read the reply and the per-call token usage](https://zato.io/docs/ai/llm/invoke.html) [### Chat - multi-turn conversations Conversations with managed history - the platform keeps each chat's turns and trims what goes to the provider](https://zato.io/docs/ai/llm/chat.html) [### Skills Reusable instructions sent as the system context of a call, shared with MCP gateways](https://zato.io/docs/ai/skills/index.html)

## Configuration and operations

[### Creating connections Providers and models, addresses, API keys, pool size, timeout, token and history limits](https://zato.io/docs/ai/llm/connections.html) [### GitOps with enmasse Export and import the whole AI configuration as YAML across dev, test and production](https://zato.io/docs/ai/enmasse.html) [### Audit log Every call is recorded with its outcome and duration](https://zato.io/docs/admin/audit-log/index.html)

## Agents calling your services

[### MCP gateways AI agents calling your Python services as tools over MCP](https://zato.io/docs/ai/mcp/index.html) [### AI gateway LLM connections and MCP gateways as the platform's AI gateway](https://zato.io/ai-gateway.html)

## Schedule a demo

A one-on-one session covers LLM connections and MCP gateways end to end.

[Schedule a demo](https://zato.io/services/demo.html)