Auto Pricing

Pricing (/pricing) under Configuration schedules cost backfills. It recalculates cost on existing LLM traces using the per-model prices in Manage models. Use it when a trace was ingested without cost, you changed prices, or you added a custom model after traffic already landed.

This page is /pricing. It is not a tab on a Costs page.

Enable Auto Pricing

1. Open Configuration → Pricing.
2. Toggle Enable Auto Pricing on.
3. Enter a cron expression (for example */15 * * * *, 0 * * * *, 0 0 * * *).
4. Click Save or Update.

On each run, ARMS finds recent LLM spans that are missing cost (or have cost 0), looks up provider and model in Manage Models, computes cost from token counts, and writes it back. Spans that already have a non-zero cost are left alone. Spans with missing provider/model/tokens, or a model that is not in Manage Models, are skipped.

Scheduled jobs require ELSAI_CRON_SECRET on the ARMS server. See Configuration.

Manual recalculation

1. Open an LLM request in Telemetry.
2. In the trace detail panel, use the Cost tile refresh icon (Recalculate cost using the model's price in Manage Models).
3. A toast shows Cost updated.

Unlike Auto Pricing, a manual recalculation refreshes cost even when a value is already present. The control only appears when the span has a model and span id.

Frequently asked questions

Why is a trace's cost missing or zero?

Usually the SDK did not send cost, or the model was not in Manage Models (or in pricing_json) when the span was ingested. Add or fix the model, then recalculate or wait for the next Auto Pricing run.

Does Auto Pricing overwrite an existing cost?

No. Auto Pricing only fills missing or zero-cost spans.