Skip to main content

Usage tracking

What it is

  • Pulls provider usage/quota directly from their usage endpoints.
  • No estimated costs; only the provider-reported windows.

Where it shows up

  • /status in chats: emoji‑rich status card with session tokens + estimated cost (API key only). Provider usage shows for the current model provider when available.
  • /usage off|tokens|full in chats: per-response usage footer (OAuth shows tokens only).
  • /usage cost in chats: local cost summary aggregated from the configured session store (JSONL transcripts or the sqlite/postgres store).
  • CLI: wednesdayai status --usage prints a full per-provider breakdown.
  • CLI: wednesdayai channels list prints the same usage snapshot alongside provider config (use --no-usage to skip).
  • macOS menu bar: “Usage” section under Context (only if available).

Per-model attribution

/usage cost and sessions.usage split the per-model rows by resolved model, not just the configured alias:
  • model is the configured alias (e.g. low); actualModel is the resolved backend model from the assistant entry’s responseModel (e.g. zai-org/GLM-5.3-Flash).
  • byModel rows are keyed provider::model::actualModel; the third segment is empty when the resolved model is unknown. One alias can therefore appear as several rows — one per resolved model — and per-row counts/tokens/cost always add back up to the session totals.
  • The Control UI usage view shows the resolved model per row, and its model filter matches against actualModel as well.

Historical backfill

On the first startup after upgrading, the session store backfills the usage and model columns for historical entries once (marker usage-columns-v1), so older entries gain per-model rows and actualModel too. The backfill runs after startup without blocking it; a failed run is logged (startup backfill usage columns failed) and retried on the next start, because the marker is written only after the last batch commits.

Providers + credentials

  • Anthropic (Claude): OAuth tokens in auth profiles.
  • GitHub Copilot: OAuth tokens in auth profiles.
  • Gemini CLI: OAuth tokens in auth profiles.
  • Antigravity: OAuth tokens in auth profiles.
  • OpenAI Codex: OAuth tokens in auth profiles (accountId used when present).
  • MiniMax: API key (coding plan key; MINIMAX_CODE_PLAN_KEY or MINIMAX_API_KEY); uses the 5‑hour coding plan window.
  • z.ai: API key via env/config/auth store.
Usage is hidden if no matching OAuth/API credentials exist.