Balance & token usage

Check your balance

Open the Gameron balance page and enter your key. It shows the current allowance, remaining tokens, expiry, and available usage records. Keep your key out of screenshots you share publicly.

Understand the token categories

A follow-up can reuse a long conversation history, so a short typed message can still process many tokens. Cached tokens can count toward your plan’s token allowance; discounted cache pricing and token totals are different measurements. Use the categories shown for each request when reviewing usage.

1. Input: new prompt content processed for the request.
2. Output: generated tokens reported for the response, including reasoning tokens where included in the model’s usage.
3. Cache creation: prompt tokens written into a cache, where reported.
4. Cache read: repeated prompt tokens served from a cache, where reported.

Ask for a usage review

Send Gameron the request reference, model, date, time and timezone, plus the token figures you want checked. Use the GM reference shown in an error or request log. Do not send private prompts or your full API key in a public channel.