The Known Good Updated 8 Oct 2026 Subscribe

Speech Arena

Elo ratings from blind pairwise votes: a human is shown two anonymous outputs for the same prompt and picks the better one. The votes are collected and published by TTS Arena — we ingest the ratings, we do not run the votes and we do not adjust them.

Text Image Video Speech

Speech Arena

Human preference rating from blind pairwise votes, ingested from TTS Arena

Text to Speech Arena

Elo from blind pairwise preference votes, collected and published by TTS Arena. Higher is better. We ingest these ratings; we do not run the votes and we do not run our own evaluations.

⇩
The Known Good
Text to Speech Arena — full ranking
46 models rated Ratings captured 8 Oct 2026 Source: TTS Arena
# Model Creator Elo 95% CI Votes Known Good Index
2 uni-tts-preview Unknown 1,625 — — —
3 Luna TTS Unknown 1,586 — — —
1 U2-TTS Unknown 1,577 — — —
3 Luck Dolphin Unknown 1,577 — — —
4 Aurora Unknown 1,566 — — —
1 Vocu V3.0 Unknown 1,564 — — —
2 CastleFlow v1.0 Unknown 1,561 — — —
7 Luck Dolphin Turbo Unknown 1,560 — — —
3 Inworld TTS MAX Unknown 1,558 — — —
5 Papla P1 Unknown 1,550 — — —
6 Inworld TTS Unknown 1,542 — — —
9 MiniMax Speech 2.8 HD MiniMax 1,540 — — —
7 star-june-2026 Unknown 1,540 — — —
21 Deepdub eTTS 3.2 Unknown 1,530 — — —
8 Hume Octave Hume AI 1,526 — — —
10 Lightning v3.1 Pro Unknown 1,526 — — —
11 Gradium TTS Unknown 1,525 — — —
14 MiniMax Speech 2.8 Turbo MiniMax 1,523 — — —
12 MiniMax Speech 2.6 Turbo MiniMax 1,521 — — —
17 OpenAudio S2 Unknown 1,520 — — —
13 MiniMax Speech 02 HD MiniMax 1,514 — — —
24 Cartesia Sonic 2 Cartesia 1,513 — — —
23 Typecast SSFM 3.0 Unknown 1,512 — — —
25 Inworld TTS 1.5 MAX Unknown 1,511 — — —
15 Eleven Flash v2.5 ElevenLabs 1,510 — — —
18 Eleven Multilingual v2 ElevenLabs 1,509 — — —
16 Eleven Turbo v2.5 ElevenLabs 1,509 — — —
26 Hithink Speech 2.6 Unknown 1,508 — — —
22 OpenAudio S1 Unknown 1,508 — — —
19 MiniMax Speech 02 Turbo MiniMax 1,505 — — —
27 MiniMax Speech 2.6 HD MiniMax 1,505 — — —
20 Parmesan Unknown 1,503 — — —
28 Eleven v3 ElevenLabs 1,501 — — —
32 Tontaube V0 Unknown 1,485 — — —
31 Voice.ai Text to Speech V1 Unknown 1,485 — — —
29 Chatterbox Unknown 1,481 — — —
30 Kokoro v1.0 Unknown 1,478 — — —
33 Magpie Research Preview Unknown 1,474 — — —
36 NLS Pre V1 Unknown 1,465 — — —
34 NeuTTS Max Unknown 1,440 — — —
36 Silk Mulberry 1.5 Unknown 1,422 — — —
38 Fluxions Vui Unknown 1,413 — — —
37 Maya 1 Unknown 1,411 — — —
35 Magpie Multilingual Unknown 1,407 — — —
39 Veena Unknown 1,365 — — —
38 Wordcab TTS Unknown 1,361 — — —
The Known Good
How to read Elo. A rating is only meaningful against the population that produced it: a model's number in one arena cannot be compared to its number in another, and a rating built on a few hundred votes moves far more than one built on tens of thousands — the 95% confidence interval is the honest guide to that. Elo measures which output a person preferred, not whether it was correct; for correctness see the evaluations and the Known Good Index, which does not include Elo. Ratings are ingested from TTS Arena and are published under the source's own terms — see attribution.