Google: Gemini 3.5 Transcribe
google/gemini-3.5-transcribe
Gemini 3.5 Transcribe is a speech-to-text model from Google. It is suited for synchronous transcription that needs word-level timestamps or speaker diarization, with support for up to eight speakers. Audio can be up to one hour, or 30 minutes when timestamps or diarization are enabled.
Input
$2
/ 1M tokens · Token reference
Output
$12
/ 1M tokens · Token reference
Context window
98.3K
tokens
Transcription output · Token reference fields do not establish an output-unit price.
Limits
Context
98,304
tokens
Max output
32,768
33% of the context window
Model facts
Author
Tokenizer
Gemini
Modality
Audio -> Transcription
Created
Sep 25, 2026
Publication status
Hugging Face: not mappedLicense: not publishedParameter count: not published
Modalities
Input · AudioOutput · Transcription