Costs

Get cost, bucketed by period and optionally grouped by one dimension (customer/feature/model/provider).

import weflayr

client = weflayr.client(api_key="wf-...")

const weflayr = require('weflayr');

const client = weflayr.client({ apiKey: 'wf-...' });

# Passed as a Bearer token on every request
export WEFLAYR_API_KEY="wf-..."

Get costs

from weflayr.openapi.models import CostsInput

client.get_costs(body=CostsInput.from_dict({
    "date_from": "2026-06-01",
    "aggregate": "week",
    "group_by": "customer",
    "customer_tags": {"region": ["EU", "NA"], "plan": "premium"},
    "run_tags": {"agent": ["orchestrator", "qa"]},
}))

await weflayr.api.getCosts({
  client,
  throwOnError: true,
  body: {
    dateFrom: '2026-06-01',
    aggregate: 'week',
    groupBy: 'customer',
    customerTags: { region: ['EU', 'NA'], plan: 'premium' },
    runTags: { agent: ['orchestrator', 'qa'] },
  },
});

curl -X POST https://app.weflayr.com/api/costs/ \
  -H "Authorization: Bearer $WEFLAYR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "date_from": "2026-06-01",
    "aggregate": "week",
    "group_by": "customer",
    "customer_tags": {"region": ["EU", "NA"], "plan": "premium"},
    "run_tags": {"agent": ["orchestrator", "qa"]}
  }'

Output:

{
  "rows": [
    {"date_from": "2026-06-01", "date_to": "2026-06-07", "customer": "c_123", "cost": 342.87, "avg_feature_run_cost": 1.42, "max_feature_run_cost": 6.10},
    {"date_from": "2026-06-01", "date_to": "2026-06-07", "customer": "c_456", "cost": 128.05, "avg_feature_run_cost": 0.85, "max_feature_run_cost": 2.30},
    {"date_from": "2026-06-08", "date_to": "2026-06-14", "customer": "c_123", "cost": 401.20, "avg_feature_run_cost": 1.55, "max_feature_run_cost": 7.40}
  ],
  "next_cursor": null,
  "has_more": false
}

Only the field named by group_by is present on each row (customer above); it’s omitted entirely when group_by is unset.

Input params: passed in the request body.

Field	Accepts	Behaviour
`date_from` Required	`"YYYY-MM-DD"`	Start of the period, inclusive.
`date_to` Optional	`"YYYY-MM-DD"`	End of the period, inclusive. Defaults to today.
`aggregate` Optional	`"day"` \| `"week"` \| `"month"` \| `"quarter"` \| `"year"` \| `"all"`	One row per period; `"all"` collapses the whole date range into a single row (per `group_by`, if set). Defaults to `"day"`.
`group_by` Optional	`"customer"` \| `"feature"` \| `"model"` \| `"provider"`	Split each period into one row per dimension value. Omit for one row per period.
`customers` Optional	list[string]	Restrict to these customers. Omit for every customer. See Get customers.
`features` Optional	list[string]	Restrict to these feature tags. Omit for every feature. See Get features.
`models` Optional	list[string]	Restrict to these models. Omit for every model. See Get models.
`providers` Optional	list[string]	Restrict to these providers. Omit for every provider. See Get providers.
`customer_tags` Optional	`{tag_key: list[string] \| string}`	Restrict to customers carrying these tags (AND across `tag_key`, OR within a `tag_key`’s values).
`run_tags` Optional	`{tag_key: list[string] \| string}`	Restrict to runs carrying these tags (AND across `tag_key`, OR within a `tag_key`’s values).
`apply_credits` Optional	boolean	Apply free AI credits discounts to `cost`. Never applied to `avg_feature_run_cost`/`max_feature_run_cost`, which always report gross cost. Defaults to `false`.
`limit` Optional	integer	Max number of rows to fetch. Defaults to `100`, max `1000`.
`cursor` Optional	string	The route is paginated: pass the previous page’s `next_cursor` to fetch the next page. PS: `has_more` from the output says whether another page exists.