Fact · Speech recognition
25audio tokens per second
Google's pricing notes estimate the cost of Gemini 3.5 Transcribe at 25 audio tokens per second of input.
Checked against its source on October 17, 2026
In the source's own words
“Estimated pricing is based on 25 audio tokens per second for input and 175 text tokens per minute for output”
Gemini Developer API pricing, Google, Google AI for Developers, 2026, Note under the Gemini 3.5 Transcribe price table.
Open the original- What it measures
- The page uses this rate to produce the per-minute estimates for the transcription model. The general audio guide gives a different rate for Gemini models in general.
- What it does not show
- This is the billing estimate for the transcription model only. The general audio guide counts 32 tokens per second, so the two rates do not match.