Skip to main content

audio model comparison

Gemini 3.5 Live Translate Preview vs Gemini 3.5 Transcribe

Gemini 3.5 Live Translate Preview and Gemini 3.5 Transcribe are both current audio models, from Google. Gemini 3.5 Transcribe is 86% cheaper per audio minute than Gemini 3.5 Live Translate Preview, $0.005 vs $0.0368.

Data verified October 13, 2026 · Dataset 2026-10-13

Share this

Side by side

Standard-tier prices in effect today. The better value in each row is highlighted.

FactGemini 3.5 Live Translate PreviewGemini 3.5 Transcribe
Price per audio minute$0.0368$0.005 (better)
Artificial Analysis AA-WER v2Not measured2.6 % WER (#4 of 6)
ProviderGoogleGoogle
Release dateNot reportedAugust 26, 2026
AccessAPI onlyAPI only
Context window131,072 tokensNot reported
Max output65,536 tokensNot reported
Inputaudioaudio
Outputtext, audiotext

Scores:Artificial Analysis · 2026-10-13

When to choose each

Gemini 3.5 Live Translate Preview

Best for

  • Realtime spoken-language translation

Watch out for

  • This preview endpoint supports translation only, rather than tools or instructions, and may change before stable release
Full Gemini 3.5 Live Translate Preview profile

Gemini 3.5 Transcribe

Best for

  • Batch transcription with speaker labels
  • Multilingual meeting and call transcripts

Watch out for

  • One hour of audio per request, or 30 minutes with diarization or timestamps
  • Custom vocabulary cannot be combined with diarization or timestamps, and there is no Batch API
Full Gemini 3.5 Transcribe profile

Gemini 3.5 Live Translate Preview vs Gemini 3.5 Transcribe FAQ

Which is cheaper, Gemini 3.5 Live Translate Preview or Gemini 3.5 Transcribe?

Gemini 3.5 Transcribe is 86% cheaper per audio minute than Gemini 3.5 Live Translate Preview, $0.005 vs $0.0368.

Production evaluation

Choosing between Gemini 3.5 Live Translate Preview and Gemini 3.5 Transcribe?

Public benchmarks narrow the shortlist. We run both on your own tasks, data and cost limits, then build the routing and review controls to run the winner in production.

  • A head-to-head evaluation on your real workload
  • Routing between models by cost and difficulty