Skip to content
Frontier Terminal
All alerts

Google cuts gemini-3.5-flash-lite batch cached input from 0.02 to 0.015 per MTok

notable

On the batch tier of gemini-3.5-flash-lite, cached input falls from 0.02 to 0.015 per MTok while input holds at 0.15 and output at 1.25. A cached-input-only cut on Google's economy line lowers the cost of repeated-context, high-volume batch workloads.

aistudio.google.com946e0b7bbaad66a961ff9f25f21a867cb9dd0ef0925c343b144bb554f6c4bbeb

Cited changes

Every field of each cited change, before and after.

modified modelgemini-3.5-flash-lite (batch) — $0.15 in / $1.25 out per Mtok

Record fields before and after the change
Field Before After
cached_input_per_mtok 0.020.015
context_window ∅∅
currency USDUSD
effective_from ∅∅
effective_until ∅∅
input_per_mtok 0.150.15
model_slug gemini-3.5-flash-litegemini-3.5-flash-lite
notes input: $0.15 (text / image / video / audio); cached_input: $0.02 | $1.00 / 1,000,000 tokens per hour (storage price)input: $0.15 (text / image / video / audio); cached_input: $0.015 | $1.00 / 1,000,000 tokens per hour (storage price)
output_per_mtok 1.251.25
provider googlegoogle
tier batchbatch
aistudio.google.commodel:google:gemini-3.5-flash-lite:batch detected

Data from the last 90 days is CC BY 4.0. Older data is CC BY-NC 4.0, except what we published before 2026-11-04, which stays CC BY 4.0. How you can reuse it.