d18d16e0ba
This commit introduces a new `InferenceTokens` billing resource type. Billing rates are added that define how much a user is billed for each inference token used. Billing depends on the amount of tokens and the type of model used for inference.
58 lines
1.8 KiB
Ruby
58 lines
1.8 KiB
Ruby
# frozen_string_literal: true
|
|
|
|
require "yaml"
|
|
|
|
class BillingRate
|
|
def self.rates
|
|
@@rates ||= YAML.load_file("config/billing_rates.yml", permitted_classes: [Time])
|
|
end
|
|
|
|
def self.from_resource_properties(resource_type, resource_family, location, active_at = Time.now)
|
|
rates.select {
|
|
_1["resource_type"] == resource_type && _1["resource_family"] == resource_family && _1["location"] == location && _1["active_from"] < active_at
|
|
}.max_by { _1["active_from"] }
|
|
end
|
|
|
|
def self.from_id(billing_rate_id)
|
|
rates.find { _1["id"] == billing_rate_id }
|
|
end
|
|
|
|
def self.line_item_description(resource_type, resource_family, amount)
|
|
case resource_type
|
|
when "VmCores"
|
|
"#{resource_family}-#{(amount * 2).to_i} Virtual Machine"
|
|
when "VmStorage"
|
|
"#{amount.to_i} GiB Storage for Virtual Machine"
|
|
when "IPAddress"
|
|
"#{resource_family} Address"
|
|
when "PostgresCores"
|
|
"#{resource_family}-#{(amount * 2).to_i} backed PostgreSQL Database"
|
|
when "PostgresStandbyCores"
|
|
"#{resource_family}-#{(amount * 2).to_i} backed PostgreSQL Database (HA Standby)"
|
|
when "PostgresStorage"
|
|
"#{amount.to_i} GiB Storage for PostgreSQL Database"
|
|
when "PostgresStandbyStorage"
|
|
"#{amount.to_i} GiB Storage for PostgreSQL Database (HA Standby)"
|
|
when "GitHubRunnerMinutes"
|
|
"#{resource_family} GitHub Runner"
|
|
when "GitHubRunnerConcurrency"
|
|
"Additional GitHub Runner Concurrency"
|
|
when "InferenceTokens"
|
|
"#{resource_family} Inference Tokens"
|
|
else
|
|
fail "BUG: Unknown resource type for line item description"
|
|
end
|
|
end
|
|
|
|
def self.line_item_usage(resource_type, resource_family, amount, duration)
|
|
case resource_type
|
|
when "GitHubRunnerMinutes"
|
|
"#{amount.to_i} minutes"
|
|
when "InferenceTokens"
|
|
"#{amount.to_i} tokens"
|
|
else
|
|
"#{duration} minutes"
|
|
end
|
|
end
|
|
end
|