List Barndoor's default pricing - Barndoor Developer Documentation

[ { "id": "3c90c3cc-0d44-4b50-8888-8dd25736052a", "provider_slug": "openai", "model_name": "gpt-4o-mini", "model_pattern": "gpt-4o-mini", "input_cost_per_million_tokens": 0.15, "output_cost_per_million_tokens": 0.6, "currency": "USD", "source": "<string>", "cache_read_cost_per_million_tokens": 123, "cache_write_cost_per_million_tokens": 123 } ]

Authorizations

Authorization

string

header

required

JWT obtained through Barndoor's authentication flow. Pass the token verbatim in Authorization: Bearer <token>. Use the Barndoor SDK's loginInteractive() helper to obtain a token in scripts and notebooks.

Response

200 - application/json

List of default pricing entries

string<uuid>

required

Stable identifier for this default row. Pass in ids to import or sync a subset.

provider_slug

string

required

Catalog slug of the provider this pricing applies to.

Example:

"openai"

model_name

string

required

Human-readable model identifier.

Example:

"gpt-4o-mini"

model_pattern

string

required

Pattern matched against the upstream model name when billing. Usually the same as model_name; may be a * wildcard for providers that bill all variants at one rate.

Example:

"gpt-4o-mini"

input_cost_per_million_tokens

number

required

USD (or currency) cost per 1M input tokens.

Example:

0.15

output_cost_per_million_tokens

number

required

USD (or currency) cost per 1M output tokens.

Example:

0.6

currency

string

required

ISO-4217 currency code. Barndoor publishes its managed defaults in USD.

Example:

"USD"

source

string

required

Provenance label for this row (e.g. provider YAML name, vendor price-sheet snapshot). Surfaced in the admin UI so operators can tell where a default came from.

cache_read_cost_per_million_tokens

number

Barndoor's managed prompt-cache read rate. Omitted when the provider/model doesn't support caching (embeddings, vision, classic text-completion, etc.) — the field is absent from the JSON rather than sent as 0 so the admin UI hides the cache row entirely instead of rendering a misleading "$0.00".

cache_write_cost_per_million_tokens

number

Barndoor's managed prompt-cache write rate. Same omission semantics as cache_read_cost_per_million_tokens.