Module: OmniAI::Anthropic::Chat::UsageSerializer

Defined in:
lib/omniai/anthropic/chat/usage_serializer.rb

Overview

Overrides usage deserialize to read Anthropic's reasoning breakdown.

Class Method Summary collapse

Class Method Details

.deserialize(data) ⇒ OmniAI::Chat::Usage

Anthropic already counts reasoning inside output_tokens and reports the breakdown alongside it at output_tokens_details.thinking_tokens, so the breakdown is read into thinking_tokens and the output count is left exactly as reported. Adding the two together would double count.

When streaming, the breakdown arrives only on the final message_delta.

Parameters:

Returns:



16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
 
# File 'lib/omniai/anthropic/chat/usage_serializer.rb', line 16

def self.deserialize(data, *)
  # Deserialize without a context so the generic flat parse runs rather than recursing into this method.
  usage = OmniAI::Chat::Usage.deserialize(data)

  # Only overwrite when the vendor container is actually present. A payload produced by `Usage#serialize`
  # carries base's own `thinking_tokens` key and no `output_tokens_details`, so assigning unconditionally
  # would clobber a correctly-parsed value with nil and break the round-trip. `unless nil?` rather than
  # `||=`, so a reported zero from the wire still wins over base's nil.
  thinking_tokens = data.dig("output_tokens_details", "thinking_tokens")
  usage.thinking_tokens = thinking_tokens unless thinking_tokens.nil?

  # Anthropic's `input_tokens` excludes cache reads and writes; adding them back keeps it the whole prompt, as
  # for every other provider. A base-serialized payload has neither key and is already whole.
  cache_tokens = data.values_at("cache_creation_input_tokens", "cache_read_input_tokens").compact
  usage.input_tokens = (usage.input_tokens || 0) + cache_tokens.sum if cache_tokens.any?

  usage
end