whatmodel.sh / stt / Gemini 3.5 Transcribe
Gemini 3.5 Transcribe
↗ share
+ contribute
Gemini 3.5 Transcribe is a closed-weights speech-to-text model from Google, released August 2026. Licence: proprietary.
Score within the Open ASR English short-form WER (March 2026) comparison set only.
$ how it compares
The speech-to-text models ranked either side of Gemini 3.5 Transcribe, out of 176 with a score, then more from Google.
| # | model | score | where | ||
|---|---|---|---|---|---|
| 9 |  Soniox v5 Real-Time | 67±3 | api $0.0020/audio min | ||
| 10 | Soniox v4 Real-Time | 67±3 | comparison only | ||
| 11 | Gemini 3.5 Transcribe | 67±6 | api $0.0050/audio min | ||
| 12 |  Scribe v2 | 66±1 | api $0.0037/audio min | ||
| 13 |  GPT Transcribe | 64±4 | api $0.0045/audio min | ||
| more from Google | |||||
| 31 |  Chirp 3 | 61±2 | api price unknown | ||
| 33 | Gemini 3.5 Transcribe Live | 61±3 | api $0.0090/audio min | ||
| 55 | Chirp 2 | 58±3 | api price unknown | ||
see all speech-to-text models · find the ones that fit your machine
$ what it's good at
other tests1 not in a task · 86% of its score
| test | result | vs the others in its run | counts |
|---|---|---|---|
| Persian ASR Triple Threat · VisualEars6669 + FLEURS-fa equal-weight WER | 6.4% ✎ | █·········██ beats 1/2 | 86% |
$ where to run it
Through an API
api · $0.0050/audio min · ai.google.dev ↗ · price sourceProvider-published estimated cost per audio minute: Google lists gemini-3.5-transcribe at $2.00/1M audio input tokens or $0.003/min and $12.00/1M text output tokens or $0.002/min, with 'an effective blended rate of ~$0.005 per min for Transcribe'; underlying billing is per token. Product: Gemini Developer API. API model: gemini-3.5-transcribe. Settings: Standard paid tier. Price basis: vendor-published per-minute estimate (25 audio tokens/s, 175 text tokens/min). Exclusions: free tier, Transcribe Live. Currency: USD. Google Gemini API pricing page (CC BY 4.0), observed 2026-09-25.
$ how the score was worked out · Persian ASR Triple Threat Lea… 86%
Each result implies a score on its own, after adjusting for how hard its test is and who ran it. The score is their weighted average, with a small pull toward the middle for thinly tested models. Each test's weight is its counts column above; here it's added up by who ran the tests. How it works
| source | results | share of the score |
|---|---|---|
 Persian ASR Triple Threat Leaderboard (Reza2kn) | 1 | 86% |
| prior · pull toward the middle | 14% | |
| weighted average | = 66.8 |
$ improve this page
Use a short GitHub form or edit the data through a pull request. Every number needs a source; here's which ones we can use.
data updated · sources and licences · json