Skip to content
IFAI

Model rankings

1542 models · 5 boards · 2 sources

‹
Week of October 12, 2026
November 9, 202636 changesNovember 2, 202641 changesOctober 26, 20269 changesOctober 19, 2026247 changesOctober 12, 2026127 changesOctober 5, 202634 changesSeptember 28, 202617 changesSeptember 21, 2026115 changesSeptember 14, 2026174 changesSeptember 7, 2026144 changesAugust 31, 202620 changesAugust 24, 202615 changes
›

Text-to-speech arena · Anonymous head-to-head votes · Updated October 13, 2026

Best voice
Eleven v4 TurboElevenLabs1,330Score
Best open model
Breeze TTS 2BreezeBlue1,223Score
Best from China
Qwen-Audio-3.1-TTS-PlusAlibaba1,298Score
Top three by Score
1Eleven v4 TurboElevenLabs1,330
2Eleven v4ElevenLabs1,326
3Qwen-Audio-3.1-TTS-PlusAlibaba1,298
Top three open models
1Breeze TTS 2BreezeBlue · #10 overall1,223
2Fish Audio S2 ProFish Audio · #33 overall1,116
3Step Audio EditX (Mar 2026)StepFun · #39 overall1,098
This week's changes
Claude Opus 5.5 (Max)Coding · #1New
Eleven v4 TurboSpeech · #1New
GPT-6 Astra (Max)Coding · #2New
Claude Sonnet 5.5 (xHigh)Coding · #3New
GPT-6.1 Sol (Max)Coding · #4New
AllOpenClosed
All regionsUSChina
One row per modelAll variants
RankChangeModelScore95% interval
1NewEleven v4 TurboClosedElevenLabs1,330±18
2▼ 1Eleven v4ClosedElevenLabs1,326±17
3▼ 1Qwen-Audio-3.1-TTS-PlusClosedAlibaba1,298±17
4▼ 1Sonic 3.6ClosedCartesia1,286±15
5▼ 1Gemini 3.8 Flash TTSClosedGoogle1,274±15
6▼ 1Qwen-Audio-3.0-TTS-PlusClosedAlibaba1,267±15
7▼ 1Realtime TTS-2ClosedInworld1,260±16
8—Gemini 3.8 Flash-Lite TTSClosedGoogle1,246±14
9▼ 2Simba 3.2ClosedSpeechifyAI1,244±13
10—Breeze TTS 2OpenBreezeBlue1,223±15
11—Realtime TTS-2 FlashClosedInworld1,219±14
12▼ 3Luna TTSClosedVUI Labs1,218±12
13—Gemini 3.1 Flash TTSClosedGoogle1,214±11
14▼ 2StepAudio 2.5 TTS (Aug 2026)ClosedStepFun1,213±15
15▼ 1v3 ConversationalClosedElevenLabs1,210±14
16▼ 1Sonic 3.5ClosedCartesia1,193±12
17▼ 1Soniox TTS Real-Time v2ClosedSoniox1,180±11
18▼ 1Eleven v3ClosedElevenLabs1,177±11
19▼ 1Speech 2.8 HDClosedMiniMax1,174±11
20NewGradium TTS (Sep 2026)ClosedGradium1,173±14
21▼ 2Lightning V3.1 Pro (Jul 2026)ClosedSmallest.ai1,172±13
22▼ 2Falcon 2ClosedMurf AI1,155±13
23▼ 1Gradium TTS (Aug 2026)ClosedGradium1,153±13
24▼ 3Speech 2.8 TurboClosedMiniMax1,153±11
25—SpaceXAI TTSClosedSpaceXAI1,145±14
26▼ 3Fish Audio S2.1 ProClosedFish Audio1,143±12
27▼ 1Step TTS 2 (Mar 2026)ClosedStepFun1,139±12
28▼ 4Async Flash v1.5ClosedAsync1,138±12
29▼ 2Azure HD 2.5ClosedMicrosoft1,134±12
30▼ 2Speech 2.6 HDClosedMiniMax1,129±11
31—Async Pro v1.0ClosedAsync1,125±12
32▼ 3Simba 3.0ClosedSpeechifyAI1,125±12
33▼ 3Speech 2.6 TurboClosedMiniMax1,124±11
34▼ 2Fish Audio S2 ProOpenFish Audio1,116±12
35▼ 2Speech-02-HDClosedMiniMax1,113±11
36▼ 1TTS-1 HDClosedOpenAI1,106±11
37▼ 1Chatterbox HDClosedResemble AI1,104±12
38▼ 4Speech 2.6ClosedHithink1,100±13
39▼ 2Turbo v2.5ClosedElevenLabs1,100±10
40▼ 1Step Audio EditX (Mar 2026)OpenStepFun1,098±12
41▼ 3Multilingual v2ClosedElevenLabs1,096±10
42▼ 1Speech-02-TurboClosedMiniMax1,091±11
43▼ 3TTS-1ClosedOpenAI1,090±11
44▼ 2Voxtral TTSOpenMistral1,088±12
45▼ 2Gemini 2.5 Flash Lite TTSClosedGoogle1,085±12
46▼ 2OpenAudio S1ClosedFish Audio1,083±11
47▼ 2StudioClosedGoogle1,081±11
48▼ 1GPT-Realtime-2ClosedOpenAI1,079±13
49▼ 3Flash v2.5ClosedElevenLabs1,078±11
50▼ 2JourneyClosedGoogle1,075±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 “—”.