Skip to main content

OpenAICartesiaAudio matchupSpeech output

GPT-4o Mini TTSvs.Sonic 3.6

Which model is better?

Sonic 3.6 does better in independent tests.

Independent tests
GPT-4o Mini TTS0 of 3
Sonic 3.6ahead1 of 3
VoiceArena, Coval, DesignArena
Popularity
GPT-4o Mini TTS36
Sonic 3.639
out of 100
Weights and access
GPT-4o Mini TTSproprietary
Sonic 3.6proprietary
Price
GPT-4o Mini TTS$0.60 per 1M text tokens
Sonic 3.6n/a
verbatim from the provider

Pick GPT-4o Mini TTS for

No clear edge

Pick Sonic 3.6 for

VoiceArenaleaning Covalleaning DesignArena

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”.

VoiceArena

Sonic 3.6 ahead

Sonic 3.6 is clearly ahead in 4 of 4 subtests, GPT-4o Mini TTS in none.

VoiceArena TTS Elo: Hindi

GPT-4o Mini TTS834
Sonic 3.61,094

95 % ranges 824 to 844 and 1,081 to 1,107, as of October 28, 2026

VoiceArena TTS Elo: Mexican Spanish

GPT-4o Mini TTS875
Sonic 3.61,081

95 % ranges 853 to 897 and 1,065 to 1,097, as of October 28, 2026

VoiceArena TTS Elo: Brazilian Portuguese

GPT-4o Mini TTS940
Sonic 3.61,078

95 % ranges 931 to 949 and 1,062 to 1,094, as of October 28, 2026

Show all 4 subtests
SubtestGPT-4o Mini TTSSonic 3.6Verdict
VoiceArena TTS Elo: Hindi8341,094Sonic 3.6 ahead
VoiceArena TTS Elo: Mexican Spanish8751,081Sonic 3.6 ahead
VoiceArena TTS Elo: Brazilian Portuguese9401,078Sonic 3.6 ahead
VoiceArena TTS Elo: US English9261,059Sonic 3.6 ahead

Source · latest October 28, 2026

Coval

Sonic 3.6 slightly ahead

Coval TTS leaderboard

GPT-4o Mini TTS2,554 ms
Sonic 3.6294 ms

lower is better, as of October 23, 2026

Source · latest October 23, 2026

DesignArena

Sonic 3.6 slightly ahead

DesignArena Audio Realism Elo

GPT-4o Mini TTS1,029
Sonic 3.61,341

as of October 28, 2026

Source · latest October 28, 2026

All data side by side

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

GPT-4o Mini TTSSonic 3.6
TaskSpeech outputSpeech output
DeveloperOpenAICartesia
StatusDeprecatedActive
Weightsproprietaryproprietary
Runs locallynono
Price$0.60 per 1M text tokensn/a
Popularity (out of 100)3639