AI Usage Tracker

Loading data

Recommended now

Loading usage…

Ranking free capacity against how soon each weekly limit resets (use-it-or-lose-it) — not who has the most % left alone.

Remaining
—
Weekly reset
—
By reset
—

Claude

Max loading
—% —
Fable
—left
Session 5h
—left

Waiting for current usage.

ChatGPT

— loading
—% —

Waiting for current usage.

Grok

— loading
—% —

Waiting for current usage.

Claude → Grok → ChatGPT Presets stay one tap away · custom 1st/2nd/3rd below · drag on desktop · saved on this device
Custom order 1st / 2nd / 3rd · drag on desktop
Show in order
Drag cards or pace panels on desktop to rearrange · clears preset highlight

Claude live pace

by reset if pace holds · sleep hours skipped
All models · by reset
—
—
Fable only · by reset
—
—

Usage trajectory

this cycle → each model’s weekly finish
Claude actual Claude projected ChatGPT Grok 100% limit
Usage trajectory for Claude, ChatGPT, and Grok

Loading the current trajectory.

Token usage

local coding-agent transcripts · activity volume — the % meters above remain the limit source of truth
—
tokens today
—
messages today
—
top model share
—
last 14 days

Daily token usage stacked by model for the last 14 days
Checkpoints ahead projected usage at upcoming checkpoints vs even pace
CheckpointUsedFableBy resetEven-paceChatGPTGrokStatus
30-min audit working-hours snapshots · newest first
DayTimeAll %Fable %By resetFable by resetvs evenHeadroom
Open to load…

↓ Download CSV

Resume clock history watch progress toward normal use · Hit 100% / Resume normal this week

Loading…

Observed Used By reset Hit 100% Resume normal Δ Resume Δ Used Note
Waiting for data…

Primary use: keep an eye on how usage is going so you know when you can get back to normal pace. Resume earlier (Δ negative) = progress; later (Δ positive) = more throttle still needed. Same week-avg math as the live clocks, recomputed from hourly checkpoints — also a clean trail if a meter step looks wrong. ↓ Download CSV