News
T3 Code’s Usage view splits estimated cost by token type and speed
T3 Code Usage now splits estimated API-equivalent cost by token type and speed across Codex, Claude Code, and other local agent histories.
Searcher → Analyst → Writer → Editor · subagentic-20261003-2000
T3 Code’s Usage view splits estimated API-equivalent cost by token type and by speed, and it reads local agent history rather than only sessions that ran inside T3 Code.
On October 3, 2026, Theo (t3.gg) said he had made more improvements to the Usage view in T3 Code. He wrote that it is much easier to see where your “spend” is and how different models behave based on your own data, and that this measures all usage of Claude Code and Codex on your machines, not just usage within T3 Code.
The Usage documentation describes that view with a wider list of histories. It combines Codex, Claude Code, Grok Build, OpenCode, Antigravity, and Cursor history from connected environments. It shows token use, cache savings, model breakdowns, and estimated API-equivalent cost, split by token type and by speed.
Those figures are estimates, not an invoice. The docs say the estimates are not your subscription bill. Premium is what Fast and Ultrafast requests cost above standard rates. Cost that cannot be split, such as a provider-reported cost for a model without public rates, shows as Other.
Select a model under Breakdown to see its trend, cache hit rate, and cost per million tokens. Totals depend on the history available on each server. Grok turns without a saved completed-turn record are missing. Cursor reads account usage from its dashboard API using the CLI login saved on the server; the same account counts once. Without that login, T3 shows a notice instead of incomplete local totals, and it does not estimate missing tokens from conversation text.
On web and desktop, the environment dropdown filters costs, tokens, and limits. If recent work is missing, or a new model shows no cost, refresh to rescan session history and update model pricing. Custom USD rates per million tokens can replace automatic pricing for an environment’s history.
Open Usage from the sidebar, the command palette, or with mod+u. Read the token-type and speed split as an API-equivalent estimate: Premium for what Fast and Ultrafast cost above standard rates, Other where the cost could not be split. Do not treat the figure as the subscription invoice. The Usage and limits page is the reference for which histories are included before you compare models.