Skip to content
IFAI

Model rankings

1239 models · 5 boards · 2 sources

‹
Week of May 25, 2026
July 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 changesMay 4, 202616 changesApril 27, 202624 changesApril 20, 202627 changes
›

Text-to-speech arena · Anonymous head-to-head votes · Updated May 31, 2026

Best voice
Gemini 3.1 Flash TTSGoogle1,214Score
Best open model
Fish Audio S2 ProFish Audio1,123Score
Best from China
Fun-Realtime-TTS-PreviewAlibaba1,193Score
Top three by Score
1Gemini 3.1 Flash TTSGoogle1,214
2Realtime TTS-2 - Research PreviewInworld1,209
3Sonic 3.5Cartesia1,203
Top three open models
1Fish Audio S2 ProFish Audio · #17 overall1,123
2Step Audio EditX (Mar 2026)StepFun · #21 overall1,111
3Voxtral TTSMistral · #35 overall1,067
This week's changes
Claude Opus 4.8 (Adaptive Reasoning, Max Effort)Intelligence · #1New
qwen3.7-max-20260517Coding · #3New
xAI Text to SpeechSpeech · #5New
Recraft V4.1 Utility ProImage · #6New
Fun-Realtime-TTS-PreviewSpeech · #6New
AllOpenClosed
All regionsUSChina
One row per modelAll variants
RankChangeModelScore95% interval
1▲ 1Gemini 3.1 Flash TTSClosedGoogle1,214±14
2▲ 6Realtime TTS-2 - Research PreviewClosedInworld1,209±14
3▼ 2Sonic 3.5ClosedCartesia1,203±15
4▼ 1Realtime TTS 1.5 MaxClosedInworld1,195±14
5NewxAI Text to SpeechClosedxAI1,194±15
6NewFun-Realtime-TTS-PreviewClosedAlibaba1,193±15
7▼ 3StepAudio 2.5 TTSClosedStepFun1,184±15
8▼ 3Eleven v3ClosedElevenLabs1,179±12
9▼ 3Speech 2.8 HDClosedMiniMax1,167±12
10▼ 3Inworld TTS 1 MaxClosedInworld1,160±13
11▼ 2Realtime TTS 1.5 MiniClosedInworld1,151±12
12▼ 2Step TTS 2 (Mar 2026)ClosedStepFun1,142±14
13▼ 1Speech 2.6 TurboClosedMiniMax1,139±11
14▼ 3Speech 2.8 TurboClosedMiniMax1,138±11
15▼ 1SIMBA 3.0ClosedSpeechify1,133±13
16▼ 3Speech 2.6 HDClosedMiniMax1,133±11
17▼ 2Fish Audio S2 ProOpenFish Audio1,123±14
18▲ 1Azure HD 2.5ClosedMicrosoft1,121±14
19▼ 3Inworld TTS 1ClosedInworld1,117±12
20▼ 3Speech-02-HDClosedMiniMax1,112±11
21▼ 3Step Audio EditX (Mar 2026)OpenStepFun1,111±14
22▼ 2Multilingual v2ClosedElevenLabs1,108±10
23▼ 2Turbo v2.5ClosedElevenLabs1,105±10
24▲ 14Gradium TTSClosedGradium1,096±15
25—Flash v2.5ClosedElevenLabs1,096±10
26▼ 4TTS-1 HDClosedOpenAI1,095±12
27▼ 1Speech-02-TurboClosedMiniMax1,094±11
28▼ 4TTS-1ClosedOpenAI1,092±10
29▼ 1Sonic 3ClosedCartesia1,085±12
30NewLightning V3.1 ProClosedSmallest.ai1,085±14
31▼ 4MiMo-V2.5-TTSClosedXiaomi1,081±14
32▼ 2Gemini 2.5 Flash Lite TTSClosedGoogle1,080±12
33▼ 4StudioClosedGoogle1,077±11
34▼ 1Polly GenerativeClosedAmazon1,069±11
35▼ 4Voxtral TTSOpenMistral1,067±14
36▲ 5Polly Long-FormClosedAmazon1,065±13
37▼ 3SIMBA 1.6ClosedSpeechify1,065±13
38▼ 3Azure NeuralClosedMicrosoft1,063±11
39▼ 3Kokoro 82M v1.0OpenKokoro1,062±10
40▼ 8T2A-01-HDClosedMiniMax1,062±11
41NewAsync Flash v1.0ClosedAsync1,061±11
42▼ 5OpenAudio S1ClosedFish Audio1,059±11
43▲ 2Chirp 3: HDClosedGoogle1,056±12
44▼ 4Magpie-Multilingual 357M (Feb 2026)OpenNVIDIA1,056±14
45▼ 1Octave 2ClosedHume AI1,054±12
46▼ 4Sonic English (Oct 2024)ClosedCartesia1,052±11
47▼ 4Maya1OpenMaya Research1,051±12
48▼ 2JourneyClosedGoogle1,046±13
49▼ 2Lightning v3.1ClosedSmallest.ai1,038±13
50—SIMBA 1.0ClosedSpeechify1,036±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 “—”.