Skip to content
IFAI

Model rankings

1412 models · 5 boards · 2 sources

Text-to-speech arena · Anonymous head-to-head votes · Updated August 12, 2026

Best voice
Simba 3.2SpeechifyAI1,231Score
Best open model
Fish Audio S2 ProFish Audio1,123Score
Best from China
Qwen-Audio-3.0-TTS-PlusAlibaba1,230Score
Top three by Score
1Simba 3.2SpeechifyAI1,231
3Luna TTSVUI Labs1,214
Top three open models
1Fish Audio S2 ProFish Audio · #23 overall1,123
2Step Audio EditX (Mar 2026)StepFun · #27 overall1,109
3Voxtral TTSMistral · #42 overall1,071
This week's changes
Qwen-Audio-3.0-TTS-PlusSpeech · #2New
Qwen3.8 MaxIntelligence · #5New
grok-4.6-highCoding · #5New
Muse Spark 1.2 (xhigh)Intelligence · #7New
gemini-3.7-flash-highCoding · #8New
RankChangeModelScore95% interval
1—Simba 3.2ClosedSpeechifyAI1,231±15
2NewQwen-Audio-3.0-TTS-PlusClosedAlibaba1,230±15
3▲ 7Luna TTSClosedVUI Labs1,214±15
4▼ 2Gemini 3.1 Flash TTSClosedGoogle1,209±13
5▲ 6StepAudio 2.5 TTSClosedStepFun1,204±15
6▼ 3Sonic 3.5ClosedCartesia1,201±13
7▼ 1Realtime TTS 1.5 MaxClosedInworld1,192±13
8NewLightning V3.1 Pro (Jul 2026)ClosedSmallest.ai1,189±16
9▼ 4Realtime TTS-2 - Research PreviewClosedInworld1,189±13
10▼ 2Speech 2.8 HDClosedMiniMax1,174±12
11▲ 1Eleven v3ClosedElevenLabs1,171±12
12NewGradium TTS BetaClosedGradium1,153±16
13▲ 1Speech 2.8 TurboClosedMiniMax1,153±12
14NewFalcon 2ClosedMurf AI1,149±17
15▼ 6Async Flash v1.5ClosedAsync1,149±14
16▲ 2Fish Audio S2.1 ProClosedFish Audio1,141±14
17▼ 1Step TTS 2 (Mar 2026)ClosedStepFun1,139±13
18▼ 5Lightning V3.1 Pro TTS (Jun 2026)ClosedSmallest.ai1,136±14
19—Speech 2.6 HDClosedMiniMax1,132±12
20▼ 3Realtime TTS 1.5 MiniClosedInworld1,128±12
21—Azure HD 2.5ClosedMicrosoft1,127±13
22▼ 7Async Pro v1.0ClosedAsync1,127±14
23▲ 1Fish Audio S2 ProOpenFish Audio1,123±13
24▼ 4Speech 2.6 TurboClosedMiniMax1,121±12
25▼ 3Simba 3.0ClosedSpeechifyAI1,120±13
26▲ 1Speech-02-HDClosedMiniMax1,111±12
27▼ 4Step Audio EditX (Mar 2026)OpenStepFun1,109±14
28▼ 21SpaceXAI TTSClosedSpaceXAI1,105±18
29▼ 1TTS-1 HDClosedOpenAI1,101±12
30▼ 5Multilingual v2ClosedElevenLabs1,100±11
31▼ 5Turbo v2.5ClosedElevenLabs1,099±11
32—Chatterbox HDClosedResemble AI1,089±13
33—TTS-1ClosedOpenAI1,085±11
34▼ 3Gradium TTSClosedGradium1,084±14
35▼ 6Flash v2.5ClosedElevenLabs1,081±11
36▼ 6Speech-02-TurboClosedMiniMax1,080±12
37▼ 1Gemini 2.5 Flash Lite TTSClosedGoogle1,080±12
38▲ 1StudioClosedGoogle1,078±12
39NewMaya 2 FlashClosedMaya Research1,073±15
40▲ 11Speech 2.6ClosedHithink1,073±14
41—OpenAudio S1ClosedFish Audio1,072±12
42▼ 8Voxtral TTSOpenMistral1,071±13
43▼ 6Sonic 3ClosedCartesia1,070±12
44▲ 6Magpie-Multilingual 357M (Feb 2026)OpenNVIDIA · 1 more variant1,066±13
45▼ 5Polly GenerativeClosedAmazon1,064±12
46▲ 7JourneyClosedGoogle1,064±13
47▼ 5SIMBA 1.6ClosedSpeechifyAI1,063±13
48▼ 10GPT-Realtime-2ClosedOpenAI1,062±14
49▼ 6T2A-01-HDClosedMiniMax1,061±11
50▼ 15MiMo-V2.5-TTSClosedXiaomi1,060±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 “—”.