> ## Documentation Index > Fetch the complete documentation index at: https://docs.moonbounce.ai/llms.txt > Use this file to discover all available pages before exploring further. # Engine and model cost for a decision, per evaluation and in total > Engine cost of one decision's evaluations and previews over a window; reads only. What the decision's evaluations cost over a window, for the key's organization and environment. Amounts are US dollars in micros (`usd_micros`: millionths of a dollar). Costs are what our engines reported, priced by each adapter from the vendor's published price, plus an estimate for calls we couldn't price from a vendor report (no usage reported, or a suggestion-writer model with no listed price), broken out as `estimated_usd_micros`; cache hits cost nothing. Evaluations that never committed and suggestion generation (its replays and model call) are in `other_spend`, outside the per-evaluation numbers, and `total` adds everything recorded. Per-request `cost` on evaluate and seeds is reported cost only. `since` (inclusive) and `until` (exclusive) are read to the millisecond and default to the last 30 days ending now; the window may span at most 366 days. A decision with no traffic in the window answers 200 with empty groups. The web session may not read it (403). ## OpenAPI ````yaml /openapi.json get /v1/decisions/{decision_id}/usage openapi: 3.1.0 info: title: Moonbounce Decisions API version: 0.0.0 description: >- Moonbounce decisions: evaluate items against a decision spec (features combined by aggregators), record feedback and seed examples, review a ranked queue, and improve the spec through suggestions that ship as code diffs, each with a preview (backtest) and an evidence label. **Authentication.** Send an API key as `Authorization: Bearer ` on every operation except `GET /health` and `POST /v1/specs/hash`. Call the API from a server, never a browser: a key must not reach one, and the API sends no CORS headers. A key belongs to one organization and one environment: production, or test for a key with the `env:test` permission. Test traffic never counts as production traffic. **Requests.** JSON bodies (`Content-Type: application/json`) of at most 1 MB, parsed strictly: duplicate keys, lone surrogates, integers outside ±(2^53 - 1), `__proto__` keys, U+0000 and nesting deeper than 64 are a 400. **Retries.** Send an `Idempotency-Key` on every POST under /v1 that takes one (all but `POST /v1/specs/hash`). Keep it for the request's retries. The same key with the same request returns the first stored result, marked `Idempotent-Replayed: true`, for 24 hours (410 if that result was since deleted); with another request it is 422 IDEMPOTENCY_KEY_REUSED. Keys are scoped to the organization and environment across every operation, so use a fresh key per request. A request that failed before its result was stored runs again when retried. 429 and 503 under /v1 may carry `Retry-After` (seconds); a daily QUOTA_EXCEEDED cap instead names its reset time. **Errors.** Every error under /v1 is `{"error": {"type", "message", "reason"?, "param"?, "details"?}}` (ErrorBody); `GET /health` answers its own shape. Every response carries a `Request-Id` header. Rate limits apply per key and per client IP, counted per Cloudflare location. servers: - url: https://api.moonbounce.ai description: Production security: [] paths: /v1/decisions/{decision_id}/usage: get: tags: - usage summary: Engine and model cost for a decision, per evaluation and in total description: >- Engine cost of one decision's evaluations and previews over a window; reads only. What the decision's evaluations cost over a window, for the key's organization and environment. Amounts are US dollars in micros (`usd_micros`: millionths of a dollar). Costs are what our engines reported, priced by each adapter from the vendor's published price, plus an estimate for calls we couldn't price from a vendor report (no usage reported, or a suggestion-writer model with no listed price), broken out as `estimated_usd_micros`; cache hits cost nothing. Evaluations that never committed and suggestion generation (its replays and model call) are in `other_spend`, outside the per-evaluation numbers, and `total` adds everything recorded. Per-request `cost` on evaluate and seeds is reported cost only. `since` (inclusive) and `until` (exclusive) are read to the millisecond and default to the last 30 days ending now; the window may span at most 366 days. A decision with no traffic in the window answers 200 with empty groups. The web session may not read it (403). operationId: getUsage parameters: - schema: $ref: '#/components/schemas/DecisionId' required: true name: decision_id in: path - schema: type: string format: date-time description: 'Start of the window, inclusive. Default: `until` - 30 days.' required: false description: 'Start of the window, inclusive. Default: `until` - 30 days.' name: since in: query - schema: type: string format: date-time description: 'End of the window, exclusive. Default: now.' required: false description: 'End of the window, exclusive. Default: now.' name: until in: query responses: '200': description: The decision's cost report content: application/json: schema: $ref: '#/components/schemas/DecisionUsage' '400': description: 'Invalid request: `param` names the field at fault when one applies' content: application/json: schema: $ref: '#/components/schemas/ErrorBody' '401': description: Missing or invalid API key (reason KEY_INVALID) content: application/json: schema: $ref: '#/components/schemas/ErrorBody' '403': description: The key may not do this content: application/json: schema: $ref: '#/components/schemas/ErrorBody' '429': description: >- RATE_LIMITED: retry after Retry-After seconds; QUOTA_EXCEEDED: details.reset_at gives the daily reset time (no Retry-After), or details.too_large asks for a smaller request headers: Retry-After: schema: type: string description: Seconds to wait before retrying, when waiting helps. required: false description: Seconds to wait before retrying, when waiting helps. content: application/json: schema: $ref: '#/components/schemas/ErrorBody' '503': description: >- Temporarily unavailable, with Retry-After (for example reason AUTH_UNAVAILABLE, ENGINE_UNAVAILABLE or STORE_UNAVAILABLE): retry with the same Idempotency-Key headers: Retry-After: schema: type: string description: Seconds to wait before retrying, when waiting helps. required: true description: Seconds to wait before retrying, when waiting helps. content: application/json: schema: $ref: '#/components/schemas/ErrorBody' security: - bearerAuth: [] components: schemas: DecisionId: type: string pattern: ^[a-z][a-z0-9_.-]{0,127}$ DecisionUsage: type: object properties: decision_id: $ref: '#/components/schemas/DecisionId' environment: $ref: '#/components/schemas/Environment' window: type: object properties: since: type: string format: date-time until: type: string format: date-time required: - since - until evaluations: type: array items: type: object properties: engine_profile: type: string assignment_mode: type: string evaluations: type: integer minimum: 0 uncached: type: integer minimum: 0 description: >- Evaluations that called an engine, answered or not; the rest were answered from the feature cache. engine_calls: type: integer minimum: 0 description: >- Upstream calls made, including ones answered with an error and ones abandoned at the deadline; `engines[]` counts every attempt. input_tokens: type: integer minimum: 0 output_tokens: type: integer minimum: 0 usd_micros: type: object properties: total: type: number minimum: 0 mean: type: number minimum: 0 p50: type: number minimum: 0 p95: type: number minimum: 0 required: - total - mean - p50 - p95 description: >- Per evaluation, in millionths of a US dollar: reported cost plus the estimate for its calls that reported no usage. estimated_usd_micros: type: number minimum: 0 description: >- The estimated part of `usd_micros.total`: calls that may have been billed but reported no usage. required: - engine_profile - assignment_mode - evaluations - uncached - engine_calls - input_tokens - output_tokens - usd_micros - estimated_usd_micros description: >- One group per engine profile and assignment mode, ordered by both. Spend whose evaluation is gone is grouped under `unknown`, as is dry-run spend, which has no evaluation. engines: type: array items: type: object properties: engine: type: string resolved_model: type: string adapter: type: string calls: type: integer minimum: 0 description: Call attempts, retries and failures included. failed: type: integer minimum: 0 description: Attempts whose status is not `ok`. input_tokens: type: integer minimum: 0 output_tokens: type: integer minimum: 0 usd_micros: type: number minimum: 0 description: Reported plus estimated. estimated_usd_micros: type: number minimum: 0 description: >- The estimated part: calls that may have been billed but reported no usage. required: - engine - resolved_model - adapter - calls - failed - input_tokens - output_tokens - usd_micros - estimated_usd_micros description: >- One group per engine, resolved model and adapter, from the per-call traces. Evaluations stored before traces existed and dry runs contribute none. other_spend: type: array items: type: object properties: kind: type: string description: >- `uncommitted`: evaluations, dry runs, seed examples and preview runs whose engines ran but which never committed (a failed evaluate, a refused seed batch, the loser of an idempotency race, a preview that failed). `generation`: suggestion generation's own replays and model call (its proposal preview is in `previews`). New kinds may appear. attempts: type: integer minimum: 0 engine_calls: type: integer minimum: 0 input_tokens: type: integer minimum: 0 output_tokens: type: integer minimum: 0 model_input_tokens: type: integer minimum: 0 description: >- The suggestion writer's language model tokens, apart from engine tokens. model_output_tokens: type: integer minimum: 0 usd_micros: type: number minimum: 0 description: Reported plus estimated. estimated_usd_micros: type: number minimum: 0 required: - kind - attempts - engine_calls - input_tokens - output_tokens - model_input_tokens - model_output_tokens - usd_micros - estimated_usd_micros description: >- Spend that is not an evaluation, one group per kind, kept out of the per-evaluation numbers. total: type: object properties: usd_micros: type: number minimum: 0 estimated_usd_micros: type: number minimum: 0 required: - usd_micros - estimated_usd_micros description: 'Everything recorded above: evaluations, other spend and previews.' previews: type: object properties: runs: type: integer minimum: 0 features_run: type: integer minimum: 0 description: >- Feature answers the runs computed (not cached); one engine call can answer several. input_tokens: type: integer minimum: 0 usd_micros: type: number minimum: 0 description: Reported plus estimated. estimated_usd_micros: type: number minimum: 0 required: - runs - features_run - input_tokens - usd_micros - estimated_usd_micros description: >- Stored preview runs in the window (`mb test`, previews, and the one preview each generated suggestion runs), which write no evaluations. Generation's exploratory replays and model call are in `other_spend`, as is a preview run that stored nothing (`uncommitted`). required: - decision_id - environment - window - evaluations - engines - other_spend - total - previews ErrorBody: type: object properties: error: type: object properties: type: $ref: '#/components/schemas/ErrorType' reason: type: string pattern: ^[A-Z][A-Z0-9_]+$ description: >- Set where a caller can act on it, one of the reasons in the ErrorReason schema. New reasons may be added: treat an unknown one by the status and `type`. message: type: string description: For people; don't parse it. param: type: string description: >- The request field at fault, as a dotted path (`subject.0.text`, `claims.1.value`). details: type: object additionalProperties: {} required: - type - message additionalProperties: false required: - error additionalProperties: false Environment: type: string enum: - production - test description: >- The environment of the key that made the call: `production` or `test`. Evaluations, feedback, reviews and suggestions are separate per environment; registered structures are shared. ErrorType: type: string enum: - invalid_request - authentication - permission_denied - not_found - conflict - rate_limited - engine_failed - unavailable - internal securitySchemes: bearerAuth: type: http scheme: bearer description: >- An API key, sent as `Authorization: Bearer `. Each operation lists the scope its key needs: `decisions:evaluate` (evaluate and read evaluations), `decisions:feedback` (your app's feedback) or `decisions:develop` (everything else, including owner feedback and the review queue). An operation listing no scope takes any scoped key. A key with no scope, or without the operation's, gets 403 INSUFFICIENT_SCOPE. ```` This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.