> ## Documentation Index > Fetch the complete documentation index at: https://docs.barndoor.ai/llms.txt > Use this file to discover all available pages before exploring further. # Create a model pricing rule ## OpenAPI ````yaml /api-reference/openapi.yaml post /api/llm-gateway/admin/model-pricing openapi: 3.1.0 info: title: Barndoor.ai Public API description: Unified public API specification for Barndoor.ai platform version: 2.18.0 servers: - url: https://{host} description: Barndoor public API variables: host: default: platform.barndoor.ai description: >- The host serving your Barndoor deployment. Defaults to production; override it for another environment or a dedicated deployment. security: - ApiKey: [] - OAuth2: - openid - profile - email tags: - name: dlp x-group: Data Control Center - name: identity x-group: Identity - name: llm-gateway x-group: LLM Gateway - name: notification x-group: Notifications - name: policy x-group: Policies - name: registry x-group: MCP Gateway - name: system-management x-group: Log Exports paths: /api/llm-gateway/admin/model-pricing: post: tags: - llm-gateway summary: Create a model pricing rule operationId: create_pricing requestBody: content: application/json: schema: $ref: '#/components/schemas/CreatePricingRequest' required: true responses: '200': description: The created record content: application/json: schema: $ref: '#/components/schemas/AdminModelPricing' '400': description: Malformed request '401': description: Missing or invalid credentials '403': description: Caller is not permitted to act on this resource in this organization '409': description: >- A pricing version already exists for this rule at that `effective_from` instant components: schemas: CreatePricingRequest: type: object required: - model_pattern - input_cost_per_million_tokens - output_cost_per_million_tokens properties: cache_read_cost_per_million_tokens: type: - number - 'null' format: double description: |- Per-million-tokens rate billed for prompt-cache reads. Omit to inherit the Barndoor default (or leave NULL if the default has none, in which case the cost calculator falls back to the full input rate). Storing NULL preserves the "no override" semantics so a future seeder update can repopulate the default rate without the org row blocking it. cache_write_cost_per_million_tokens: type: - number - 'null' format: double description: |- Per-million-tokens rate billed for the cache-creation premium that Anthropic-family models charge on the request that populates the prompt cache. Same NULL semantics as `cache_read_cost`. catalog_slug: type: - string - 'null' change_reason: type: - string - 'null' description: |- Optional free-text note ("Q1 pricing review", "Negotiated discount") surfaced on the History dialog timeline. effective_from: type: - string - 'null' format: date-time description: |- When the price should take effect. Omit (or pass `null`) for "right now". A timestamp in the future schedules the change and records the row with `change_source = 'admin_schedule'` so the History dialog can distinguish it from an immediate edit. input_cost_per_million_tokens: type: number format: double long_context: oneOf: - $ref: '#/components/schemas/LongContextTierRequest' description: >- Optional context-tiered surcharge (BCP-3904). Omit for the normal case of one rate set at every prompt size. - type: 'null' model_pattern: type: string model_provider: type: - string - 'null' output_cost_per_million_tokens: type: number format: double provider_id: type: - string - 'null' format: uuid sync_mode: $ref: '#/components/schemas/PricingSyncMode' AdminModelPricing: allOf: - $ref: '#/components/schemas/ModelPricing' - type: object required: - sync_mode - change_source - is_archived properties: change_reason: type: - string - 'null' change_source: $ref: '#/components/schemas/PricingChangeSource' created_by_email: type: - string - 'null' created_by_name: type: - string - 'null' created_by_user_id: type: - string - 'null' format: uuid is_archived: type: boolean description: >- True for tombstone rows produced by `archive_rule` (V35). When the latest effective version of a group has `is_archived = true`, the rule is considered archived: the cost calculator skips it and the admin summary list hides it by default. Past non-archived versions remain in the timeline so the History dialog can show the rule's full life including the archive event. sync_mode: $ref: '#/components/schemas/PricingSyncMode' description: >- Admin-API response shape for a single pricing row. Wraps the `ai_governance::ModelPricing` used by the cost calculator so that the hot path stays sync-mode-agnostic, while the admin surface still gets the metadata it needs to render badges, the Review-updates dialog, and (as of BCP-2830 v2 / V34) the per-row history timeline. LongContextTierRequest: type: object description: >- The long-context surcharge on a create/update request. One nested object rather than five sibling fields so a partial tier is awkward to express in the first place, and so the wire shape matches both the provider YAML's `long_context:` block and the `long_context` object the defaults listing returns. required: - threshold_prompt_tokens - input_cost_per_million_tokens - output_cost_per_million_tokens properties: cache_read_cost_per_million_tokens: type: - number - 'null' format: double description: Omit to bill long-context cache reads at this tier's input rate. cache_write_cost_per_million_tokens: type: - number - 'null' format: double input_cost_per_million_tokens: type: number format: double output_cost_per_million_tokens: type: number format: double threshold_prompt_tokens: type: integer format: int64 description: |- Total prompt tokens (non-cached input + cache-read + cache-creation) at or above which the tier's rates replace the base rates for the whole request. PricingSyncMode: type: string description: |- Per-row sync behavior. Stored as a `VARCHAR(16)` constrained to `tracking | pinned | auto` (see migration V33). enum: - tracking - pinned - auto ModelPricing: type: object required: - id - org_id - model_pattern - input_cost_per_million_tokens - output_cost_per_million_tokens - effective_from properties: cache_read_cost_per_million_tokens: type: - number - 'null' format: double description: |- Per-million-token cost for input tokens served from the prompt cache. `None` ⇒ the calculator bills cached input at `input_cost_per_million_tokens`. cache_write_cost_per_million_tokens: type: - number - 'null' format: double description: |- Per-million-token cost for input tokens added to the prompt cache. `None` ⇒ same fallback as `cache_read_cost_per_million_tokens`. catalog_slug: type: - string - 'null' description: >- Catalog slug for provider-level pricing (e.g. "openai", "aws_bedrock"). effective_from: type: string format: date-time id: type: string format: uuid input_cost_per_million_tokens: type: number format: double long_context: oneOf: - $ref: '#/components/schemas/LongContextTier' description: >- Optional context-tiered surcharge (BCP-3904). `None` ⇒ one rate set at every prompt size, i.e. exactly the pre-BCP-3904 behaviour. Serialized as `long_context`, not as the Rust field name. This struct reaches the admin API through `AdminModelPricing`'s `#[serde(flatten)]`, so the field name here *is* the wire key — and every sibling surface (`DefaultPricingEntry`, `EffectivePricing`, the create/update request bodies) already spells it `long_context`. Left unrenamed, the pricing tab would write a tier successfully and then read back nothing, which is precisely how it behaved before this attribute existed: the surcharge saved, the dialog reopened unchecked, and the edit looked silently discarded. Pinned by `admin_model_pricing_serializes_the_tier_as_long_context`. - type: 'null' model_pattern: type: string model_provider: type: - string - 'null' org_id: type: string format: uuid output_cost_per_million_tokens: type: number format: double provider_id: type: - string - 'null' format: uuid description: Optional reference to a specific LLM Gateway provider instance. provider_name: type: - string - 'null' PricingChangeSource: type: string description: |- How a `model_pricing` version was produced. Persisted as a `VARCHAR(20)` constrained to one of these eight values (see migrations V34 + V35). Surfaced to admins as the "source" badge on each entry of the `PricingHistoryDialog`. enum: - admin_create - admin_edit - admin_schedule - import - sync_manual - sync_auto - admin_archive - admin_restore LongContextTier: type: object description: >- The alternate rate set a pricing row switches to once a request's prompt reaches `threshold_prompt_tokens` (BCP-3904). Crossing the threshold re-rates the **whole** request, not just the tokens past it — that is how xAI, OpenAI and Azure all publish it, and it is why this is a rate switch rather than a marginal bracket. Note the asymmetry that makes this easy to get wrong: the gate is measured on the *prompt*, but the surcharge applies to output tokens too. The gate quantity and the billed quantities are different numbers. required: - threshold_prompt_tokens - input_cost_per_million_tokens - output_cost_per_million_tokens properties: cache_read_cost_per_million_tokens: type: - number - 'null' format: double description: |- `None` ⇒ bill cache reads at this tier's input rate, mirroring the base-rate fallback rather than falling back to the *base* cache rate (which would price a long-context cache read below its own tier). cache_write_cost_per_million_tokens: type: - number - 'null' format: double description: '`None` ⇒ same fallback as `cache_read_cost_per_million_tokens`.' input_cost_per_million_tokens: type: number format: double output_cost_per_million_tokens: type: number format: double threshold_prompt_tokens: type: integer format: int64 description: |- Total prompt tokens at or above which this tier replaces the base rates. Inclusive — see [`ModelPricing::rates_for_prompt`]. securitySchemes: ApiKey: type: http scheme: bearer description: >- A Barndoor API key, sent as `Authorization: Bearer `. Keys are prefixed `bdai_` and are opaque — do not attempt to decode one. Interchangeable with an OAuth 2.0 access token. OAuth2: type: oauth2 x-issuer: https://auth.barndoor.ai/realms/barndoor description: >- An OAuth 2.0 access token from Barndoor's identity provider, sent as `Authorization: Bearer `. Public clients must use PKCE. Interchangeable with an API key. flows: authorizationCode: authorizationUrl: >- https://auth.barndoor.ai/realms/barndoor/protocol/openid-connect/auth tokenUrl: >- https://auth.barndoor.ai/realms/barndoor/protocol/openid-connect/token scopes: openid: Issue an OIDC ID token identifying the caller. profile: Read the caller's basic profile claims. email: Read the caller's email address. offline_access: Issue a refresh token. Optional — no operation requires it. clientCredentials: tokenUrl: >- https://auth.barndoor.ai/realms/barndoor/protocol/openid-connect/token scopes: openid: Issue an OIDC ID token identifying the caller. profile: Read the caller's basic profile claims. email: Read the caller's email address. offline_access: Issue a refresh token. Optional — no operation requires it. ```` This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.