Skip to content

Open-source Grok Bot alternative for named AI teammates. Local-first desktop, web, and mobile clients with per-bot tools, instructions, and memory. Independent fork of T3 Code.

3.5 KB
Raw Blame

Review usage

Open Usage from the sidebar to review subscription limits and recent model activity across your connected environments. Usage opens as a workspace page, so it has its own URL and works with normal back and forward navigation.

Plan limits

The Limits tab opens by default and shows available limits for ChatGPT, Claude, Grok, Kimi For Coding, and OpenCode Go. Disconnected accounts and limits that a provider does not report stay hidden.

Each meter shows the percentage used or left and the next reset time. Provider plans expose different windows, such as five-hour, weekly, or plan-wide limits.

Akeru asks each provider for fresh limits at most once every five minutes. If a provider does not answer, the meters keep their last reported values and Akeru tries again after one minute.

Cost and tokens

Use the Cost and Tokens tabs to review activity Akeru recorded through providers connected in Settings > Providers. Choose the past 24 hours, 7 days, 30 days, or 90 days. The page shows provider totals, an hourly or daily chart, processed token totals, estimated cost, and breakdowns by model or time.

Akeru never imports machine-wide Claude, Codex, or other CLI history. Each environment reports only its own Akeru provider activity, and disconnected providers stay hidden. A notice identifies environments that are still loading, stale, or unavailable.

Per-bot usage keeps estimated model cost separate from subscription limits. A cost estimate is based on the model rate table and is not money spent. Subscription pool meters are provider-reported percentages; when a provider does not expose a meter, the value is shown as unavailable. If a provider later corrects the cache token breakdown for a turn, the cost estimate updates without counting the turn's tokens again. Tool executions and routine runs are recorded as separate zero-token lifecycle entries because those boundaries do not expose a separate provider bill. Their model tokens remain in the surrounding turn. Recall and extraction are internal memory stages without a separately billed provider boundary, so they remain covered by the parent observer or reflector entry.

Web and desktop charts stay still when idle. Hover over a chart to inspect a value.

Per-bot usage

On mobile, Settings > General > Usage lists the bots in each connected environment below the plan limits. Open a bot to see its input, output, Observer, and Reflector tokens, an estimated cost, subscription pool use, and reserved tokens. Archived bots remain in the list with an archived label so you can review their earlier usage.

A measurement a provider did not report reads Unavailable rather than zero, and a measurement that is only partly reported shows a trailing + so the number reads as a floor. A notice names the snapshot as incomplete when any provider measurement is missing.

Estimated cost comes from the model rate table. It is not subscription spend and not an amount billed. Subscription pool use is provider-reported as either a percentage or a token count; when a provider reports no meter, the row reads Unavailable.

Per-bot usage reads when you open it, when the app returns to the foreground, and when you pull to refresh. It does not poll in the background.

Refresh usage

Akeru refreshes usage every five minutes by default. Select the refresh button to reload it now. The page remembers the last tab and time range you selected.

Change Usage refresh in Settings > General to a value from 1 to 60 minutes. Mobile also shows Usage under General settings and supports pull-to-refresh.