↑↓ navigate open Esc close
Documentation

Set up your organization's AI

Choose between Torneyo AI (plan quota) or your own provider key, turn features on and off, set the tone of voice and track usage.

The settings live under Organization → AI settings (or /organizacoes/{code}/configuracao-ia). Only the organization’s owner and administrators can access them.

AI source: Torneyo or your own key

Torneyo AI (default) uses the platform key and consumes your plan’s monthly token quota. Nothing to set up.

My own key uses an AI provider you contract yourself. Advantages: the plan quota is not consumed, you pick the model and the cost is paid directly to the provider. To enable it:

  1. Select Use my own key.
  2. Choose the provider: Anthropic (Claude) or OpenAI-compatible (OpenAI, Google Gemini, Groq, NVIDIA, Ollama and any service with the same API). The base URL is filled in for the common ones; adjust it if your provider uses another.
  3. Paste the API key and enter the model (for example claude-sonnet-5 or gpt-4.1-mini).
  4. Click Test connection. Torneyo makes a short call and shows the model and latency, or the error returned by the provider (invalid key, unknown model).
  5. Save.

The key is stored encrypted and never shown again: the screen only displays its last 4 characters. To replace it, paste a new one; to keep it, leave the field empty.

Features and tone of voice

Four switches enable or disable each feature for the whole organization: help assistant, suggestions in forms, match content and Actions. The tone of voice (journalistic, casual, formal, motivational) applies to every generated text and can be overridden per championship.

The Actions switch controls the features where AI carries out tasks with your confirmation — conversation mode, voice score entry, pasting a roster from WhatsApp and creating a championship from your rulebook. Turning it off keeps the assistant and text suggestions working normally; only task execution becomes unavailable. See AI that acts: voice, rulebook and roster.

Usage and quota

The card at the top shows the source in effect, the tokens used this month and the quota, with a progress bar. The quota belongs to the plan (Plans, limits and upgrade): Free 50k, Starter 250k, Pro 2 million and Enterprise 5 million (default, adjustable by contract). No plan has unlimited AI. Past 80% the Torneyo team is notified; at 100% the features stop until the month turns, unless you use your own key.

Enterprise: quota and limits by contract

The Enterprise plan includes 5 million AI tokens per month by default, but the monthly quota and hourly request limit can be adjusted by contract to fit the organization’s needs, set and applied by the Torneyo team. The value in effect appears on the usage card as contract quota when it differs from the default.

Good practices and privacy

  • Prefer your own key if your organization already has an AI provider contract or needs a specific model.
  • Revoke the key at the provider if it leaks; in Torneyo just paste the new one.
  • When AI is enabled, operational data (teams, scores, typed text) is sent to the chosen provider. Personal data is masked before sending. See Privacy and Data Protection.

Updated on

Keep reading