ccusage / pricing-fetcher / PricingFetcher

Class: PricingFetcher

Fetches and caches model pricing information from LiteLLM Implements Disposable pattern for automatic resource cleanup

Implements

• Disposable

Constructors

Constructor

new PricingFetcher(offline): PricingFetcher;

Creates a new PricingFetcher instance

Parameters

Parameter	Type	Default value	Description
offline	boolean	false	Whether to use pre-fetched pricing data instead of fetching from API

Returns

PricingFetcher

Methods

[dispose]()

dispose: void;

Implements Disposable interface for automatic cleanup

Returns

void

Implementation of

Disposable.[dispose]

---

clearCache()

clearCache(): void;

Clears the cached pricing data

Returns

void

---

fetchModelPricing()

fetchModelPricing(): Promise<Map<string, {
  input_cost_per_token?: number;
  output_cost_per_token?: number;
  cache_creation_input_token_cost?: number;
  cache_read_input_token_cost?: number;
}>>;

Fetches all available model pricing data

Returns

Promise<Map<string, { input_cost_per_token?: number; output_cost_per_token?: number; cache_creation_input_token_cost?: number; cache_read_input_token_cost?: number; }>>

Map of model names to pricing information

---

getModelPricing()

getModelPricing(modelName): Promise<
  | null
  | {
  input_cost_per_token?: number;
  output_cost_per_token?: number;
  cache_creation_input_token_cost?: number;
  cache_read_input_token_cost?: number;
}>;

Gets pricing information for a specific model with fallback matching Tries exact match first, then provider prefixes, then partial matches

Parameters

Parameter	Type	Description
modelName	string	Name of the model to get pricing for

Returns

Promise< | null | { input_cost_per_token?: number; output_cost_per_token?: number; cache_creation_input_token_cost?: number; cache_read_input_token_cost?: number; }>

Model pricing information or null if not found

---

calculateCostFromTokens()

calculateCostFromTokens(tokens, modelName): Promise<number>;

Calculates the cost for given token usage and model

Parameters

Parameter	Type	Description
tokens	{ input_tokens: number; output_tokens: number; cache_creation_input_tokens?: number; cache_read_input_tokens?: number; }	Token usage breakdown
tokens.input_tokens	number	Number of input tokens
tokens.output_tokens	number	Number of output tokens
tokens.cache_creation_input_tokens?	number	Number of cache creation tokens
tokens.cache_read_input_tokens?	number	Number of cache read tokens
modelName	string	Name of the model used

Returns

Promise<number>

Total cost in USD

---

calculateCostFromPricing()

calculateCostFromPricing(tokens, pricing): number;

Calculates cost from token usage and pricing information

Parameters

Parameter	Type	Description
tokens	{ input_tokens: number; output_tokens: number; cache_creation_input_tokens?: number; cache_read_input_tokens?: number; }	Token usage breakdown
tokens.input_tokens	number	Number of input tokens
tokens.output_tokens	number	Number of output tokens
tokens.cache_creation_input_tokens?	number	Number of cache creation tokens
tokens.cache_read_input_tokens?	number	Number of cache read tokens
pricing	{ input_cost_per_token?: number; output_cost_per_token?: number; cache_creation_input_token_cost?: number; cache_read_input_token_cost?: number; }	Model pricing rates
pricing.input_cost_per_token?	number	-
pricing.output_cost_per_token?	number	-
pricing.cache_creation_input_token_cost?	number	-
pricing.cache_read_input_token_cost?	number	-

Returns

number

Total cost in USD