CostUsage

Token counts for one model call. Each is a whole number, at least 0.

cacheCreationInputTokens
Type: integer
min:
0

Input tokens written to the prompt cache, priced at the model's input rate.
cacheReadInputTokens
Type: integer
min:
0

Input tokens read from the prompt cache, priced at the model's cache-read rate.
inputTokens
Type: integer
min:
0

Input tokens, not counting cache reads or cache writes.
outputTokens
Type: integer
min:
0

Output tokens, including any reasoning tokens.
reasoningOutputTokens
Type: integer
min:
0

How many of outputTokens were reasoning tokens. Kept with the record; priced as part of outputTokens.

SignalExportFile