2026-07-02 · Thamodharan Ganesan
If you've used TextMate AI for a while, you may have noticed your usage limit doesn't reset at a fixed time each day. That's intentional — TextMate AI uses a rolling 4-hour window, not a daily quota.
The first message you send starts a 4-hour timer. Every message you send within that window counts against your limit. Once 4 hours have passed since your first message in the window, the window resets and you get a fresh allowance — regardless of what time of day it is. This is the same mechanic several other major AI chat products use for their own usage limits, rather than a simple "resets at midnight" quota.
In practice, this means:
TextMate AI measures usage in tokens (roughly, chunks of text — both what you send and what the model generates count toward the total):
| Plan | Tokens per 4-hour window |
|---|---|
| Free | 500,000 |
| Pro | 1,000,000 |
Pro gives you 2x the token budget per window, on top of access to more advanced models and a larger context window. See the Pricing page for the full plan breakdown, or Understanding Usage Limits & Plans in the docs for a shorter reference version of this same mechanic.
A short "what's 2+2?" message costs far fewer tokens than a request that pastes in 500 lines of code and asks for a full refactor with explanation. Measuring usage in tokens (rather than a flat per-message count) means your limit reflects actual compute used, not just how many times you hit send.
Once you've used your window's full token budget, you'll need to wait for the window to roll over (up to 4 hours from your first message in that window) before sending more messages — or upgrade to Pro for a 2x larger budget. TextMate AI's Models page also breaks down which models are available on Free vs Pro, since some heavier models are Pro-only.