Documentation

Get Started

Ready to build with sovereign AI inference? Sign up to get your API key.

API Access openapi.json

The TrustedTokens API is OpenAI-compatible: point any OpenAI SDK or client at the base URL below using your account token. Streaming, tool calling, and structured outputs are supported.

Base URL

URL
https://api.trustedtokens.eu/v1

Authentication

All requests require a valid Bearer token from your TrustedTokens account. Include it in the Authorization header.

Header
Authorization: Bearer <your-trustedtokens-token>

Endpoints

MethodPathDescriptionCompatibility
POST/v1/chat/completionsCreate a chat completionOpenAI,
POST/v1/embeddingsCreate embeddingsOpenAI,
POST/v1/systemoneEvaluate state with a decision modelTypesafe,
POST/v1/responsesCreate a responseOpenAI,
POST/anthropic/v1/messagesCreate a messageAnthropic,
GET/v1/modelsList available modelsOpenAI,

The Compatibility column shows which API specification an endpoint follows. OpenAI means the request and response shapes match the OpenAI API. OpenRouter means the endpoint follows the OpenRouter API, which extends the OpenAI model schema with pricing, architecture, and reasoning metadata. Anthropic means the endpoint follows the Anthropic Messages API. Each model can be reached through the surfaces its capabilities support; embedding and decision models use /v1/embeddings and /v1/systemone respectively.

Code Examples

cURL
curl https://api.trustedtokens.eu/v1/chat/completions \
  -H "Authorization: Bearer <your-trustedtokens-token>" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "zai-org/GLM-5.3",
    "messages": [{"role": "user", "content": "Hello, TrustedTokens!"}]
  }'

Agent Config

Connect your agent of choice to TrustedTokens. Set TRUSTEDTOKENS_API_KEY in your shell profile (e.g. .bashrc or .zshrc), then paste the snippet below into the agent’s config file.