ai.internal.total_input_tokens

All input tokens, for a provider whose own input count (`uncached`) excludes cache reads and writes: null when it reported no input count.

Reference version

Signature

function ai.internal.total_input_tokens(

uncached: int | null,

cached: int | null,

cache_write: int | null

) -> int | null throws never

All input tokens, for a provider whose own input count (uncached) excludes cache reads and writes: null when it reported no input count.

Source:<builtin>/ai/ns_internal/helpers.bamlbytes 6726–6940