> ## Documentation Index > Fetch the complete documentation index at: https://www.c1.ai/docs/llms.txt > Use this file to discover all available pages before exploring further. # Get Inference Usage Attribution Rollup > Returns inference request and token usage across the Edges the caller manages, grouped by the selected attribution dimension. ## OpenAPI ````yaml https://spec.speakeasy.com/conductor-one/conductorone/my-source-with-code-samples get /api/v1/edges/inference/usage-attribution openapi: 3.1.0 info: description: The C1 API is a HTTP API for managing C1 resources. title: C1 API version: 0.1.0-alpha servers: - description: The C1 API server for the current tenant. url: https://{tenantDomain}.conductor.one variables: tenantDomain: default: example description: The domain of the tenant to use for this request. security: - bearerAuth: [] oauth: [] paths: /api/v1/edges/inference/usage-attribution: get: tags: - Edge summary: Get Inference Usage Attribution Rollup description: >- Returns inference request and token usage across the Edges the caller manages, grouped by the selected attribution dimension. operationId: c1.api.edge.v1.EdgeService.GetInferenceUsageAttributionRollup parameters: - in: query name: dimension schema: description: The dimension field. enum: - TB_USAGE_ATTRIBUTION_DIMENSION_UNSPECIFIED - TB_USAGE_ATTRIBUTION_DIMENSION_TEAM - TB_USAGE_ATTRIBUTION_DIMENSION_APPLICATION - TB_USAGE_ATTRIBUTION_DIMENSION_AGENT - TB_USAGE_ATTRIBUTION_DIMENSION_COST_CENTER - TB_USAGE_ATTRIBUTION_DIMENSION_USER - TB_USAGE_ATTRIBUTION_DIMENSION_PROVIDER - TB_USAGE_ATTRIBUTION_DIMENSION_MODEL - TB_USAGE_ATTRIBUTION_DIMENSION_ROUTE - TB_USAGE_ATTRIBUTION_DIMENSION_CALL_ROLE type: string x-speakeasy-unknown-values: allow - in: query name: page_size schema: description: The pageSize field. format: int32 type: integer - in: query name: since schema: description: The since field. format: int64 type: string - in: query name: until schema: description: The until field. format: int64 type: string responses: '200': content: application/json: schema: $ref: >- #/components/schemas/c1.api.edge.v1.EdgeServiceGetInferenceUsageAttributionRollupResponse description: Successful response components: schemas: c1.api.edge.v1.EdgeServiceGetInferenceUsageAttributionRollupResponse: description: The EdgeServiceGetInferenceUsageAttributionRollupResponse message. properties: edgeCount: description: >- Number of authorized Edges included. Zero means the caller manages no Edges, as distinct from Edges that have no traffic yet. format: uint32 type: integer rows: description: |- Ordered by (input_tokens + output_tokens) descending, limited to page_size. items: $ref: '#/components/schemas/c1.api.edge.v1.TBUsageAttributionRow' type: - array - 'null' title: Edge Service Get Inference Usage Attribution Rollup Response type: object x-speakeasy-name-override: EdgeServiceGetInferenceUsageAttributionRollupResponse c1.api.edge.v1.TBUsageAttributionRow: description: TBUsageAttributionRow is one attribution-id group. properties: attributionId: description: >- Empty string means unattributed for a user-facing attribution dimension. For PROVIDER it is the raw provider_id, and for MODEL it is the raw provider-native served model_name. Blank remains distinct from a literal "Unknown" ID; clients label it Unknown provider/model. type: string cacheReadTokens: description: >- Observed provider-attempt cache usage, not inferred for missing attempts. format: uint64 type: string cacheWriteTokens: description: The cacheWriteTokens field. format: uint64 type: string costNanoUsd: description: |- List-price estimate for observed usage in this group, in nano-USD. Provider/model usage is priced per served model; long-context rates use average input per observed attempt. Missing usage or an unpriced model makes this a lower bound, not an invoiced amount. format: int64 type: string deniedRequestCount: description: >- Logical requests denied by policy, counted in request_count. Set for the logical-request dimensions (not PROVIDER, MODEL or CALL_ROLE). format: uint64 type: string inputTokens: description: >- PROVIDER and MODEL values include only provider-reported per-attempt usage. format: uint64 type: string judgeAttemptCount: description: >- Judge (call_role auxiliary) provider attempts that reached a provider on behalf of this group's requests. Only set for the USER and ROUTE dimensions. Judge usage is extra usage on top of the answer: it is never included in input_tokens, output_tokens or cost_nano_usd above. format: uint64 type: string judgeCostNanoUsd: description: >- List-price estimate of the observed judge usage, in nano-USD; a lower bound when the judge model is unpriced or an attempt reported no usage. format: int64 type: string judgeInputTokens: description: >- Provider-reported judge usage; judge attempts with no reported usage are not counted here and make the judge totals a lower bound. format: uint64 type: string judgeMissingUsageCount: description: Judge attempts that reported no token usage. format: uint64 type: string judgeOutputTokens: description: The judgeOutputTokens field. format: uint64 type: string missingUsageCallCount: description: |- Provider attempts with contact started but no observed token usage. Their cost is unknown even when the model has a list price. format: uint64 type: string modelBreakdown: description: >- Complete breakdown of this provider's included models, only for PROVIDER. Parent counters include these children; do not sum both for scope totals. items: $ref: '#/components/schemas/c1.api.edge.v1.TBUsageAttributionModelRow' type: - array - 'null' outputTokens: description: The outputTokens field. format: uint64 type: string requestCount: description: >- Logical root requests except for PROVIDER and MODEL, where this counts contacted provider attempts. format: uint64 type: string retryAttemptCount: description: >- Provider attempts beyond the first for the same request and call role, ranked by event time rather than by the logged attempt ordinal. A judge attempt followed by its answer attempt is not a retry. Only set for the CALL_ROLE dimension. format: uint64 type: string unpricedRequestCount: description: >- Observed requests or provider attempts whose model has no list price. format: uint64 type: string title: Tb Usage Attribution Row type: object x-speakeasy-name-override: TBUsageAttributionRow c1.api.edge.v1.TBUsageAttributionModelRow: description: |- One provider-native served model within a PROVIDER attribution group. Blank model_name is distinct from a model literally named "Unknown". properties: cacheReadTokens: description: The cacheReadTokens field. format: uint64 type: string cacheWriteTokens: description: The cacheWriteTokens field. format: uint64 type: string costNanoUsd: description: The costNanoUsd field. format: int64 type: string inputTokens: description: The inputTokens field. format: uint64 type: string missingUsageCallCount: description: The missingUsageCallCount field. format: uint64 type: string modelName: description: The modelName field. type: string outputTokens: description: The outputTokens field. format: uint64 type: string requestCount: description: The requestCount field. format: uint64 type: string unpricedRequestCount: description: The unpricedRequestCount field. format: uint64 type: string title: Tb Usage Attribution Model Row type: object x-speakeasy-name-override: TBUsageAttributionModelRow securitySchemes: bearerAuth: scheme: bearer type: http oauth: description: >- This API uses OAuth2 with the Client Credential flow. Client Credentials must be sent in the BODY, not the headers. For an example of how to implement this, refer to the [c1TokenSource.Token()](https://github.com/ConductorOne/conductorone-sdk-go/blob/3375fe7c0126d17e7ec4e711693dee7b791023aa/token_source.go#L101-L187) function. flows: clientCredentials: scopes: {} tokenUrl: /auth/v1/token type: oauth2 ````