Skip to content
IFAI

Model rankings

1275 models · 5 boards · 2 sources

‹
Week of June 8, 2026
July 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 changesMay 4, 202616 changes
›

Text-to-speech arena · Anonymous head-to-head votes · Updated June 14, 2026

Best voice
Fun-Realtime-TTSAlibaba1,226Score
Best open model
Fish Audio S2 ProFish Audio1,117Score
Best from China
Fun-Realtime-TTSAlibaba1,226Score
Top three by Score
1Fun-Realtime-TTSAlibaba1,226
2Gemini 3.1 Flash TTSGoogle1,215
3Realtime TTS-2 - Research PreviewInworld1,208
Top three open models
1Fish Audio S2 ProFish Audio · #21 overall1,117
2Step Audio EditX (Mar 2026)StepFun · #25 overall1,107
3Voxtral TTSMistral · #37 overall1,077
This week's changes
Claude Fable 5 (Adaptive Reasoning, Max Effort, Opus 4.8 Fallback)Intelligence · #1New
claude-fable-5Coding · #1New
HiDream-O1-Image-1.5Image · #3New
MAI-Image-2.5-Flash (Foundry)Image · #8New
MAI-Image-2.5 (Foundry)Image · #12New
AllOpenClosed
All regionsUSChina
One row per modelAll variants
RankChangeModelScore95% interval
1—Fun-Realtime-TTSClosedAlibaba1,226±16
2—Gemini 3.1 Flash TTSClosedGoogle1,215±13
3▲ 2Realtime TTS-2 - Research PreviewClosedInworld1,208±14
4—Sonic 3.5ClosedCartesia1,197±15
5▼ 2xAI Text to SpeechClosedxAI1,196±14
6—Realtime TTS 1.5 MaxClosedInworld1,195±14
7—Async Flash v1.5ClosedAsync1,181±15
8—StepAudio 2.5 TTSClosedStepFun1,177±16
9▲ 1Eleven v3ClosedElevenLabs1,176±12
10▼ 1Speech 2.8 HDClosedMiniMax1,170±12
11▲ 1Inworld TTS 1 MaxClosedInworld1,161±13
12▼ 1Async Pro v1.0ClosedAsync1,159±15
13▲ 2Step TTS 2 (Mar 2026)ClosedStepFun1,155±14
14—Speech 2.8 TurboClosedMiniMax1,152±12
15▼ 2Realtime TTS 1.5 MiniClosedInworld1,151±12
16—Speech 2.6 TurboClosedMiniMax1,137±11
17—Speech 2.6 HDClosedMiniMax1,136±12
18—SIMBA 3.0ClosedSpeechify1,129±13
19▲ 1Azure HD 2.5ClosedMicrosoft1,121±13
20▲ 1Inworld TTS 1ClosedInworld1,121±12
21▼ 2Fish Audio S2 ProOpenFish Audio1,117±14
22▲ 2Turbo v2.5ClosedElevenLabs1,109±10
23—Speech-02-HDClosedMiniMax1,108±12
24▲ 1Multilingual v2ClosedElevenLabs1,107±10
25▼ 3Step Audio EditX (Mar 2026)OpenStepFun1,107±14
26—Gradium TTSClosedGradium1,098±15
27▲ 5TTS-1 HDClosedOpenAI1,096±12
28▼ 1Flash v2.5ClosedElevenLabs1,095±11
29▼ 1Lightning V3.1 ProClosedSmallest.ai1,090±14
30—Speech-02-TurboClosedMiniMax1,090±12
31▼ 2TTS-1ClosedOpenAI1,086±10
32▼ 1Chatterbox HDClosedResemble AI1,086±13
33▲ 1Sonic 3ClosedCartesia1,083±12
34▼ 1Gemini 2.5 Flash Lite TTSClosedGoogle1,082±12
35—StudioClosedGoogle1,081±11
36▲ 1MiMo-V2.5-TTSClosedXiaomi1,078±14
37▼ 1Voxtral TTSOpenMistral1,077±14
38—Polly GenerativeClosedAmazon1,072±12
39—Azure NeuralClosedMicrosoft1,067±12
40▲ 2Polly Long-FormClosedAmazon1,064±13
41▼ 1T2A-01-HDClosedMiniMax1,064±11
42▲ 1Kokoro 82M v1.0OpenKokoro1,061±11
43▼ 2OpenAudio S1ClosedFish Audio1,060±11
44▲ 4SIMBA 1.6ClosedSpeechify1,060±13
45▲ 2Magpie-Multilingual 357M (Feb 2026)OpenNVIDIA · 1 more variant1,060±14
46▼ 1Octave 2ClosedHume AI1,060±12
47▼ 3Async Flash v1.0ClosedAsync1,059±11
48▼ 2Chirp 3: HDClosedGoogle1,056±12
49▲ 4Maya1OpenMaya Research1,051±12
50▲ 1JourneyClosedGoogle1,050±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 “—”.