Skip to content
IFAI

Model rankings

1512 models · 5 boards · 2 sources

‹
Week of September 21, 2026
November 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 changesAugust 17, 202633 changes
›

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

Best voice
Sonic 3.6Cartesia1,273Score
Best open model
Breeze TTS 2BreezeBlue1,204Score
Best from China
Qwen-Audio-3.0-TTS-PlusAlibaba1,259Score
Top three by Score
1Sonic 3.6Cartesia1,273
2Gemini 3.8 Flash TTSGoogle1,260
3Qwen-Audio-3.0-TTS-PlusAlibaba1,259
Top three open models
1Breeze TTS 2BreezeBlue · #9 overall1,204
2Fish Audio S2 ProFish Audio · #27 overall1,120
3Step Audio EditX (Mar 2026)StepFun · #36 overall1,094
This week's changes
claude-opus-5.5-maxCoding · #1New
Gemini 3.8 Flash TTSSpeech · #2New
Grok Imagine Image 2.0Image · #4New
gpt-6-sol-maxCoding · #5New
Gemini 3.8 Flash-Lite TTSSpeech · #6New
AllOpenClosed
All regionsUSChina
One row per modelAll variants
RankChangeModelScore95% interval
1—Sonic 3.6ClosedCartesia1,273±17
2NewGemini 3.8 Flash TTSClosedGoogle1,260±17
3—Qwen-Audio-3.0-TTS-PlusClosedAlibaba1,259±17
4▼ 2Realtime TTS-2ClosedInworld1,245±18
5▼ 1Simba 3.2ClosedSpeechifyAI1,237±14
6NewGemini 3.8 Flash-Lite TTSClosedGoogle1,235±16
7▼ 2Luna TTSClosedVUI Labs1,230±14
8NewRealtime TTS-2 FlashClosedInworld1,210±15
9▼ 2Breeze TTS 2OpenBreezeBlue1,204±16
10▼ 1Gemini 3.1 Flash TTSClosedGoogle1,199±12
11▼ 1StepAudio 2.5 TTS (Aug 2026)ClosedStepFun1,199±17
12▼ 4v3 ConversationalClosedElevenLabs1,196±15
13▼ 2Sonic 3.5ClosedCartesia1,185±12
14▼ 2Lightning V3.1 Pro (Jul 2026)ClosedSmallest.ai1,175±14
15▼ 2Soniox TTS Real-Time v2ClosedSoniox1,173±12
16▼ 1Speech 2.8 HDClosedMiniMax1,168±11
17▼ 3Eleven v3ClosedElevenLabs1,167±11
18▼ 2Falcon 2ClosedMurf AI1,157±15
19▼ 2Speech 2.8 TurboClosedMiniMax1,149±11
20▼ 2Gradium TTS (Aug 2026)ClosedGradium1,148±14
21▼ 1Fish Audio S2.1 ProClosedFish Audio1,139±13
22▼ 1Step TTS 2 (Mar 2026)ClosedStepFun1,131±13
23▲ 1Azure HD 2.5ClosedMicrosoft1,128±12
24▲ 3SpaceXAI TTSClosedSpaceXAI1,128±15
25▼ 6Async Flash v1.5ClosedAsync1,128±13
26▼ 4Speech 2.6 HDClosedMiniMax1,127±12
27▼ 2Fish Audio S2 ProOpenFish Audio1,120±13
28—Simba 3.0ClosedSpeechifyAI1,118±12
29▼ 3Speech 2.6 TurboClosedMiniMax1,114±11
30▼ 1Async Pro v1.0ClosedAsync1,113±13
31▼ 1Speech-02-HDClosedMiniMax1,111±12
32▼ 1TTS-1 HDClosedOpenAI1,099±12
33▲ 2Chatterbox HDClosedResemble AI1,096±13
34▲ 4Speech 2.6ClosedHithink1,096±13
35▼ 2Turbo v2.5ClosedElevenLabs1,095±11
36▼ 4Step Audio EditX (Mar 2026)OpenStepFun1,094±13
37▼ 3Multilingual v2ClosedElevenLabs1,092±10
38▼ 2TTS-1ClosedOpenAI1,086±11
39—Speech-02-TurboClosedMiniMax1,085±12
40▼ 3Gemini 2.5 Flash Lite TTSClosedGoogle1,081±12
41—OpenAudio S1ClosedFish Audio1,077±11
42▲ 1Voxtral TTSOpenMistral1,076±13
43▼ 3StudioClosedGoogle1,075±12
44▼ 2Flash v2.5ClosedElevenLabs1,074±11
45▼ 1JourneyClosedGoogle1,073±13
46▲ 1GPT-Realtime-2ClosedOpenAI1,072±13
47▼ 2Maya 2 FlashClosedMaya Research1,066±14
48▲ 1Magpie-Multilingual 357M (Feb 2026)OpenNVIDIA1,063±13
49▲ 3Kokoro 82M v1.0OpenKokoro1,061±11
50▼ 2T2A-01-HDClosedMiniMax1,061±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 “—”.