Skip to main content

GoogleAssemblyAIAudio matchupSpeech to text

Gemini 3.5 Transcribe Livevs.Universal-3.6 Pro Realtime

Which model is better?

Universal-3.6 Pro Realtime is slightly ahead in independent tests, but the lead is not certain.

Independent tests
Gemini 3.5 Transcribe Live0 of 1
Universal-3.6 Pro Realtime0 of 1
Coval
Popularity
Gemini 3.5 Transcribe Liveahead35
Universal-3.6 Pro Realtime0
out of 100
Weights and access
Gemini 3.5 Transcribe Liveproprietary
Universal-3.6 Pro Realtimeproprietary
Price
Gemini 3.5 Transcribe Live$3.50 per 1M audio tokens ($0.005 per audio minute)
Universal-3.6 Pro Realtime$0.45 per hour of audio; volume discounts available
verbatim from the provider

Pick Gemini 3.5 Transcribe Live for

more popular

Pick Universal-3.6 Pro Realtime for

leaning Coval

Shared tests

Independent evaluators and leaderboards only, the newest shared snapshot per evaluator. With 95 % ranges a lead counts when they do not overlap; for hit rates, when it holds in a 95 % test. Without any uncertainty figure at most “slightly ahead”.

Coval

Universal-3.6 Pro Realtime slightly ahead

Coval Word Error Rate (WER) per-sample mean

Gemini 3.5 Transcribe Live3.4 %
Universal-3.6 Pro Realtime2.5 %

lower is better, as of November 2, 2026

Source · latest November 2, 2026

All data side by side

Prices are verbatim from the provider, because audio models bill per minute, character or token.

Gemini 3.5 Transcribe LiveUniversal-3.6 Pro Realtime
TaskSpeech to textSpeech to text
DeveloperGoogleAssemblyAI
StatusActiveActive
Releasedn/aSeptember 29, 2026
Weightsproprietaryproprietary
Runs locallynono
Price$3.50 per 1M audio tokens ($0.005 per audio minute)$0.45 per hour of audio; volume discounts available
Popularity (out of 100)350