Per-modality token counts inside a RealtimeUsage update. All fields optional — providers report different subsets (OpenAI GA: text/audio/cached on input, text/audio on output).
Optional
Audio-modality tokens.
Tokens served from the provider's prompt cache (billed at the cached rate).
Image-modality tokens (input only on current providers).
Text-modality tokens.
Per-modality token counts inside a RealtimeUsage update. All fields optional — providers report different subsets (OpenAI GA: text/audio/cached on input, text/audio on output).