Tenzro Labs
Tutorials

Build with the Tenzro Labs Platform, step by step.

Six tutorials, each with code you can copy and run. All of them use one Platform API key, starting tzl_.
Tutorials
01

Call a model with the API

Create a Platform API key, pick a model by id, and send a chat completion with curl, the OpenAI Python SDK and the OpenAI JavaScript SDK, streamed or whole.

Model API
02

Embeddings, forecasts and images

Each model kind has its own endpoint. Embed texts for search, forecast a time series with quantile bands, and call the image endpoint.

Model API
03

Give an agent the models over MCP

Connect Claude Code, Cursor or VS Code to the Platform MCP server, then call the chat, embed and forecast tools.

MCP server
04

Create an agent

Propose an agent, set what it may do and spend, and approve it with your passkey on your Tenzro Labs account. Change its limits or revoke it the same way. On the Platform, with the Platform API or with the MCP tools.

Platform API · MCP server
05

Run an agent on its own key

Give an approved agent a Platform API key and call models with it. Its spend counts against the limits you approved, and anything beyond them is refused.

Model API · Platform API
06

Manage the Platform programmatically

List models and machines, create and revoke keys, read the credit balance and activity. Every call shown with the Platform API and its MCP tool.

Platform API · MCP server
The three surfaces
  • Model API
    https://api.tenzro.xyz/v1

    OpenAI-compatible. Chat completions, embeddings, forecasts, image generations.

  • Platform API
    https://platform.tenzro.xyz/api

    Catalogue, account, credits, keys and agents, as JSON over HTTPS.

  • MCP server
    https://platform.tenzro.xyz/api/mcp

    The models and the Platform as tools for Claude Code, Cursor, VS Code and any MCP client.

Get a key on the PlatformRead the docs