Skip to content
IFAI

Model rankings

1278 models · 5 boards · 2 sources

‹
Week of June 15, 2026
July 27, 202639 changesJuly 20, 202618 changesJuly 13, 202611 changesJuly 6, 202626 changesJune 29, 202641 changesJune 22, 202612 changesJune 15, 2026184 changesJune 8, 202624 changesJune 1, 202622 changesMay 25, 202617 changesMay 18, 202616 changesMay 11, 202665 changes
›

Text-to-speech arena · Anonymous head-to-head votes · Updated June 21, 2026

Best voice
Fun-Realtime-TTSAlibaba1,224Score
Best open model
Fish Audio S2 ProFish Audio1,117Score
Best from China
Fun-Realtime-TTSAlibaba1,224Score
Top three by Score
1Fun-Realtime-TTSAlibaba1,224
2Gemini 3.1 Flash TTSGoogle1,222
3Realtime TTS-2 - Research PreviewInworld1,212
Top three open models
1Fish Audio S2 ProFish Audio · #20 overall1,117
2Step Audio EditX (Mar 2026)StepFun · #22 overall1,115
3Voxtral TTSMistral · #34 overall1,081
This week's changes
glm-5.2 (max)Coding · #2New
GLM-5.2 (max)Intelligence · #6New
seed-2.1-pro-previewCoding · #6New
Lightning V3.1 Pro TTS (Jun 2026)Speech · #11New
kimi-k2.7-codeCoding · #17New
AllOpenClosed
All regionsUSChina
One row per modelAll variants
RankChangeModelScore95% interval
1—Fun-Realtime-TTSClosedAlibaba1,224±16
2—Gemini 3.1 Flash TTSClosedGoogle1,222±13
3—Realtime TTS-2 - Research PreviewClosedInworld1,212±13
4—Sonic 3.5ClosedCartesia1,203±15
5—xAI Text to SpeechClosedxAI1,200±14
6—Realtime TTS 1.5 MaxClosedInworld1,198±13
7—Async Flash v1.5ClosedAsync1,186±15
8—StepAudio 2.5 TTSClosedStepFun1,181±16
9—Eleven v3ClosedElevenLabs1,181±12
10—Speech 2.8 HDClosedMiniMax1,174±12
11NewLightning V3.1 Pro TTS (Jun 2026)ClosedSmallest.ai1,172±16
12—Async Pro v1.0ClosedAsync1,165±15
13—Step TTS 2 (Mar 2026)ClosedStepFun1,159±14
14—Speech 2.8 TurboClosedMiniMax1,158±12
15—Realtime TTS 1.5 MiniClosedInworld1,158±12
16—Speech 2.6 TurboClosedMiniMax1,141±11
17—Speech 2.6 HDClosedMiniMax1,141±12
18—SIMBA 3.0ClosedSpeechify1,132±13
19—Azure HD 2.5ClosedMicrosoft1,130±13
20▲ 1Fish Audio S2 ProOpenFish Audio1,117±14
21▲ 3Multilingual v2ClosedElevenLabs1,115±10
22▲ 3Step Audio EditX (Mar 2026)OpenStepFun1,115±14
23▼ 1Turbo v2.5ClosedElevenLabs1,112±10
24▼ 1Speech-02-HDClosedMiniMax1,109±11
25▲ 1Gradium TTSClosedGradium · 1 more variant1,104±14
26▲ 2Flash v2.5ClosedElevenLabs1,099±11
27—TTS-1 HDClosedOpenAI1,099±12
28▲ 2Speech-02-TurboClosedMiniMax1,094±12
29▲ 2TTS-1ClosedOpenAI1,091±10
30▲ 3Sonic 3ClosedCartesia1,090±12
31▲ 1Chatterbox HDClosedResemble AI1,088±13
32▲ 2Gemini 2.5 Flash Lite TTSClosedGoogle1,088±12
33▲ 2StudioClosedGoogle1,086±11
34▲ 3Voxtral TTSOpenMistral1,081±14
35▲ 1MiMo-V2.5-TTSClosedXiaomi1,081±14
36▲ 2Polly GenerativeClosedAmazon1,076±12
37▲ 2Azure NeuralClosedMicrosoft1,071±12
38▲ 5OpenAudio S1ClosedFish Audio1,068±11
39▲ 2T2A-01-HDClosedMiniMax1,068±11
40▲ 6Octave 2ClosedHume AI1,067±12
41▼ 1Polly Long-FormClosedAmazon1,067±13
42—Kokoro 82M v1.0OpenKokoro1,066±11
43▲ 2Magpie-Multilingual 357M (Feb 2026)OpenNVIDIA · 1 more variant1,065±14
44—SIMBA 1.6ClosedSpeechify1,065±13
45▲ 2Async Flash v1.0ClosedAsync1,061±11
46▲ 2Chirp 3: HDClosedGoogle1,060±12
47▲ 3JourneyClosedGoogle1,058±13
48▲ 1Maya1OpenMaya Research1,056±12
49▲ 2Sonic English (Oct 2024)ClosedCartesia1,049±12
50▲ 2CodaClosedRime1,048±13

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 “—”.