API Access

Integrate Steinkauz into existing software and workflows through OpenAI-compatible HTTP endpoints: the same models, provider configuration, and Budget rules as chat.

Platform capabilities

  • OpenAI-compatible endpoints
  • API keys for your active organization. Owners and admins create keys.
  • Each key has a minimum provider trust level. Only configured providers and models above that threshold are available; anything below is blocked before inference.
  • Unified usage and activity log with per-key attribution

Who it's for

  • Automation and internal tools that need reliable multi-provider inference
  • Coding agents and orchestration layers that speak OpenAI-compatible APIs
  • Existing codebases using the OpenAI SDK: change the credentials and use Steinkauz within minutes
  • Organizations that want both a chat UI for humans and an API for systems

API-specific visibility

  • Per-key attribution so you know which system or workflow used budget
  • Provider trust per key: a minimum environment class (E-floor), separate from the chat routing matrix
  • API keys can set a minimum provider environment class

Full activity and administrative audit details are under Governance.

Ecosystem

Built for the OpenAI-compatible ecosystem

OpenAI-compatible endpoints are the de facto standard for agents, automation, and custom AI systems. Connect them to Steinkauz for the same models and providers as chat.

OpenCode

Coding agent with OpenAI-compatible providers and custom base URL.

n8n

Automation workflows that call chat completions through a custom OpenAI endpoint.

Hermes

Agent runtime that accepts any OpenAI-compatible inference endpoint.

Paperclip

Orchestrate agent teams; route inference through an OpenAI-compatible adapter.

And many more

The ecosystem grows every day. If a client speaks OpenAI-compatible APIs, it can use Steinkauz.

Build your own

Ship OpenAI-compatible systems on Steinkauz using the OpenAI SDK or any HTTP client as your foundation.

Quick start

Quick start

Create an API key in the platform, then call the endpoints with any OpenAI-compatible client.

  1. Sign up and subscribe to a Steinkauz plan.
  2. In the platform, go to Settings → API Keys and create a key.
  3. Point your client at the Steinkauz base URL and send a chat completion request.
# Steinkauz Platform API example (OpenAI-compatible HTTP API)

# List models and pick any. For this example, we're picking the first entry of the list.
MODEL=$(curl -sS "$STEINKAUZ_BASE_URL/v1/models" \
  -H "Authorization: Bearer $STEINKAUZ_API_KEY" | jq -r '.data[0].id')

# Send a chat completion
curl -sS "$STEINKAUZ_BASE_URL/v1/chat/completions" \
  -H "Authorization: Bearer $STEINKAUZ_API_KEY" \
  -d "{\"model\": \"$MODEL\", \"messages\": [{\"role\": \"user\", \"content\": \"Hello 👋\"}]}"

Build on a governed foundation

Use the same models, providers, and policies through OpenAI-compatible APIs.

Explore governanceRead the docs