Model rankings
1278 models · 5 boards · 2 sources
Best voice
Best open model
Best from China
Top three by Score
Fun-Realtime-TTSAlibaba1,224
Gemini 3.1 Flash TTSGoogle1,222
Realtime TTS-2 - Research PreviewInworld1,212
Top three open models
Fish Audio S2 ProFish Audio · #20 overall1,117
Step Audio EditX (Mar 2026)StepFun · #22 overall1,115
Voxtral TTSMistral · #34 overall1,081
This week's changes
glm-5.2 (max)Coding · #2New
GLM-5.2 (max)Intelligence · #6New
seed-2.1-pro-previewCoding · #6New
Lightning V3.1 Pro TTS (Jun 2026)Speech · #11New
kimi-k2.7-codeCoding · #17New
| Rank | Change | Model | Score | 95% interval |
|---|---|---|---|---|
| 1 | — | Fish Audio S2 ProOpenFish Audio | 1,117 | ±14 |
| 2 | — | Step Audio EditX (Mar 2026)OpenStepFun | 1,115 | ±14 |
| 3 | — | Voxtral TTSOpenMistral | 1,081 | ±14 |
| 4 | — | Kokoro 82M v1.0OpenKokoro | 1,066 | ±11 |
| 5 | — | Magpie-Multilingual 357M (Feb 2026)OpenNVIDIA | 1,065 | ±14 |
| 6 | — | Maya1OpenMaya Research | 1,056 | ±12 |
| 7 | — | Fish Speech 1.5OpenFish Audio | 1,013 | ±11 |
| 8 | — | Magpie-Multilingual 357MOpenNVIDIA | 1,011 | ±11 |
| 9 | — | ChatterboxOpenResemble AI | 1,006 | ±11 |
| 10 | — | Zonos-v0.1OpenZyphra | 1,000 | ±0 |
| 11 | — | VibeVoice 7BOpenMicrosoft | 970 | ±13 |
| 12 | — | OpenVoice v2OpenOpenVoice | 966 | ±13 |
| 13 | — | XTTS v2OpenCoqui | 918 | ±15 |
| 14 | — | StyleTTS 2OpenStyleTTS | 891 | ±15 |
| 15 | — | MetaVoice v1OpenMetaVoice | 837 | ±18 |
Intelligence over time
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