Skip to content
IFAI

Model rankings

1227 models · 5 boards · 2 sources

‹
Week of May 18, 2026
June 29, 202641 changesJune 22, 202612 changesJune 15, 2026184 changesJune 8, 202624 changesJune 1, 202622 changesMay 25, 202617 changesMay 18, 202616 changesMay 11, 202665 changesMay 4, 202616 changesApril 27, 202624 changesApril 20, 202627 changesApril 13, 202637 changes
›

Text-to-speech arena · Anonymous head-to-head votes · Updated May 24, 2026

Best voice
Sonic 3.5Cartesia1,218Score
Best open model
Fish Audio S2 ProFish Audio1,124Score
Best from China
StepAudio 2.5 TTSStepFun1,187Score
Top three by Score
1Sonic 3.5Cartesia1,218
2Gemini 3.1 Flash TTSGoogle1,209
3Realtime TTS 1.5 MaxInworld1,194
Top three open models
1Fish Audio S2 ProFish Audio · #15 overall1,124
2Step Audio EditX (Mar 2026)StepFun · #18 overall1,112
3Voxtral TTSMistral · #31 overall1,070
This week's changes
Sonic 3.5Speech · #1New
Qwen3.7 MaxIntelligence · #5New
Gemini 3.5 Flash (high)Intelligence · #6New
grok-imagine-image-qualityImage · #6New
gemini-3.5-flashCoding · #7New
AllOpenClosed
All regionsUSChina
One row per modelAll variants
RankChangeModelScore95% interval
1NewSonic 3.5ClosedCartesia1,218±16
2—Gemini 3.1 Flash TTSClosedGoogle1,209±14
3▼ 2Realtime TTS 1.5 MaxClosedInworld1,194±14
4▼ 1StepAudio 2.5 TTSClosedStepFun1,187±15
5▼ 1Eleven v3ClosedElevenLabs1,184±12
6—Speech 2.8 HDClosedMiniMax1,168±12
7▼ 2Inworld TTS 1 MaxClosedInworld1,164±13
8NewRealtime TTS-2 - Research PreviewClosedInworld1,158±14
9▼ 1Realtime TTS 1.5 MiniClosedInworld1,158±12
10▼ 1Step TTS 2 (Mar 2026)ClosedStepFun1,150±14
11▼ 1Speech 2.8 TurboClosedMiniMax1,145±11
12—Speech 2.6 TurboClosedMiniMax1,140±11
13▼ 2Speech 2.6 HDClosedMiniMax1,137±11
14▼ 7SIMBA 3.0ClosedSpeechify1,136±13
15▼ 2Fish Audio S2 ProOpenFish Audio1,124±14
16—Inworld TTS 1ClosedInworld1,120±12
17▼ 2Speech-02-HDClosedMiniMax1,117±11
18—Step Audio EditX (Mar 2026)OpenStepFun1,112±14
19▼ 5Azure HD 2.5ClosedMicrosoft1,110±14
20▼ 3Multilingual v2ClosedElevenLabs1,110±10
21▼ 1Turbo v2.5ClosedElevenLabs1,107±10
22▼ 1TTS-1 HDClosedOpenAI1,095±11
23NewSeed-TTS 2.0ClosedByteDance Seed1,095±15
24▼ 5TTS-1ClosedOpenAI1,094±10
25▼ 1Flash v2.5ClosedElevenLabs1,092±10
26▼ 4Speech-02-TurboClosedMiniMax1,091±11
27▼ 4MiMo-V2.5-TTSClosedXiaomi1,090±14
28▼ 2Sonic 3ClosedCartesia1,082±11
29▲ 1StudioClosedGoogle1,077±11
30▼ 5Gemini 2.5 Flash Lite TTSClosedGoogle1,076±12
31▲ 4Voxtral TTSOpenMistral1,070±14
32—T2A-01-HDClosedMiniMax1,067±11
33—Polly GenerativeClosedAmazon1,066±11
34▼ 3SIMBA 1.6ClosedSpeechify1,063±13
35▲ 3Azure NeuralClosedMicrosoft1,063±11
36▼ 2Kokoro 82M v1.0OpenKokoro1,063±10
37▼ 8OpenAudio S1ClosedFish Audio1,062±11
38▼ 11Gradium TTSClosedGradium1,061±14
39▼ 3AsyncFlow V2, asyncClosedAsync1,060±11
40▼ 12Magpie-Multilingual 357M (Feb 2026)OpenNVIDIA · 1 more variant1,060±14
41▼ 1Polly Long-FormClosedAmazon1,057±13
42▼ 3Sonic English (Oct 2024)ClosedCartesia1,054±11
43▼ 6Maya1OpenMaya Research1,053±11
44▼ 3Octave 2ClosedHume AI1,052±12
45▼ 3Chirp 3: HDClosedGoogle1,048±11
46▼ 2JourneyClosedGoogle1,044±13
47▼ 2Lightning v3.1ClosedSmallest.ai1,040±13
48▼ 2Gemini 2.5 Flash TTS (Dec 2025)ClosedGoogle1,035±12
49▲ 1Octave TTSClosedHume AI1,032±11
50▼ 3SIMBA 1.0ClosedSpeechify1,032±11

Intelligence over time

Each maker's best model at the time on the Artificial Analysis Intelligence Index, by the month it was released.

Gained most · Anthropic +57.2Highest now · Anthropic 64.6Gap between highest and lowest 18.1 ptsTime range Nov 2025 – Nov 2026
64.6Anthropic60.0Google56.9OpenAI50.1Meta46.5Kimi

Arena data comes from the leaderboard dataset it publishes on Hugging Face (CC BY 4.0). Only each board's overall leaderboard is used; scores and ranks are unchanged. dataset · CC BY 4.0

Artificial Analysis data comes from its API (artificialanalysis.ai). Scores are unchanged; ranks follow the scores. artificialanalysis.ai

Open or closed follows the license Arena gives; prices and Artificial Analysis intelligence show only where model names match, otherwise “—”.