模型排行
1260 个模型 · 5 个榜单 · 2 个数据来源
语音王座
开源之光
国产之光
开源前三甲
Fish Audio S2 ProFish Audio · 总第 19 名1,122
Step Audio EditX (Mar 2026)StepFun · 总第 22 名1,112
Voxtral TTSMistral · 总第 36 名1,080
这一期的变化
Fun-Realtime-TTS语音榜 · 第 1新上榜
claude-opus-4-8-thinking编程榜 · 第 2新上榜
Cosmos3-Super-Text2Image (agentic)图像榜 · 第 5新上榜
minimax-m3编程榜 · 第 6新上榜
Krea 2 Medium图像榜 · 第 7新上榜
| 排名 | 变化 | 模型 | 得分 | 置信区间 |
|---|---|---|---|---|
| 1 | 新 | Fun-Realtime-TTS闭源Alibaba | 1,231 | ±17 |
| 2 | ▼ 1 | Gemini 3.1 Flash TTS闭源Google | 1,229 | ±15 |
| 3 | ▲ 2 | xAI Text to Speech闭源xAI | 1,208 | ±17 |
| 4 | ▼ 1 | Sonic 3.5闭源Cartesia | 1,204 | ±16 |
| 5 | ▼ 3 | Realtime TTS-2 - Research Preview闭源Inworld | 1,204 | ±15 |
| 6 | ▼ 2 | Realtime TTS 1.5 Max闭源Inworld | 1,201 | ±14 |
| 7 | 新 | Async Flash v1.5闭源Async | 1,186 | ±17 |
| 8 | ▼ 1 | StepAudio 2.5 TTS闭源StepFun | 1,181 | ±16 |
| 9 | — | Speech 2.8 HD闭源MiniMax | 1,175 | ±12 |
| 10 | ▼ 2 | Eleven v3闭源ElevenLabs | 1,173 | ±12 |
| 11 | 新 | Async Pro v1.0闭源Async | 1,165 | ±16 |
| 12 | ▼ 2 | Inworld TTS 1 Max闭源Inworld | 1,158 | ±13 |
| 13 | ▼ 2 | Realtime TTS 1.5 Mini闭源Inworld | 1,152 | ±13 |
| 14 | — | Speech 2.8 Turbo闭源MiniMax | 1,151 | ±12 |
| 15 | ▼ 3 | Step TTS 2 (Mar 2026)闭源StepFun | 1,148 | ±14 |
| 16 | ▼ 3 | Speech 2.6 Turbo闭源MiniMax | 1,142 | ±12 |
| 17 | ▼ 1 | Speech 2.6 HD闭源MiniMax | 1,136 | ±12 |
| 18 | ▼ 3 | SIMBA 3.0闭源Speechify | 1,132 | ±14 |
| 19 | ▼ 2 | Fish Audio S2 Pro开源Fish Audio | 1,122 | ±14 |
| 20 | ▼ 2 | Azure HD 2.5闭源Microsoft | 1,121 | ±14 |
| 21 | ▼ 2 | Inworld TTS 1闭源Inworld | 1,120 | ±13 |
| 22 | ▼ 1 | Step Audio EditX (Mar 2026)开源StepFun | 1,112 | ±15 |
| 23 | ▼ 3 | Speech-02-HD闭源MiniMax | 1,110 | ±12 |
| 24 | ▼ 1 | Turbo v2.5闭源ElevenLabs | 1,105 | ±11 |
| 25 | ▼ 3 | Multilingual v2闭源ElevenLabs | 1,103 | ±11 |
| 26 | ▼ 2 | Gradium TTS闭源Gradium | 1,099 | ±16 |
| 27 | ▼ 2 | Flash v2.5闭源ElevenLabs | 1,099 | ±11 |
| 28 | ▲ 2 | Lightning V3.1 Pro闭源Smallest.ai | 1,091 | ±15 |
| 29 | ▼ 1 | TTS-1闭源OpenAI | 1,090 | ±11 |
| 30 | ▼ 3 | Speech-02-Turbo闭源MiniMax | 1,090 | ±12 |
| 31 | ▲ 24 | Chatterbox HD闭源Resemble AI | 1,088 | ±14 |
| 32 | ▼ 6 | TTS-1 HD闭源OpenAI | 1,086 | ±12 |
| 33 | ▼ 1 | Gemini 2.5 Flash Lite TTS闭源Google | 1,085 | ±13 |
| 34 | ▼ 5 | Sonic 3闭源Cartesia | 1,082 | ±12 |
| 35 | ▼ 2 | Studio闭源Google | 1,080 | ±12 |
| 36 | ▼ 1 | Voxtral TTS开源Mistral | 1,080 | ±15 |
| 37 | ▼ 6 | MiMo-V2.5-TTS闭源Xiaomi | 1,078 | ±15 |
| 38 | ▼ 4 | Polly Generative闭源Amazon | 1,069 | ±12 |
| 39 | ▼ 1 | Azure Neural闭源Microsoft | 1,067 | ±12 |
| 40 | — | T2A-01-HD闭源MiniMax | 1,063 | ±11 |
| 41 | ▲ 1 | OpenAudio S1闭源Fish Audio | 1,062 | ±11 |
| 42 | ▼ 6 | Polly Long-Form闭源Amazon | 1,060 | ±14 |
| 43 | ▼ 4 | Kokoro 82M v1.0开源Kokoro | 1,059 | ±11 |
| 44 | ▼ 3 | Async Flash v1.0闭源Async | 1,059 | ±12 |
| 45 | — | Octave 2闭源Hume AI | 1,058 | ±12 |
| 46 | ▼ 3 | Chirp 3: HD闭源Google | 1,055 | ±12 |
| 47 | ▼ 3 | Magpie-Multilingual 357M (Feb 2026)开源NVIDIA | 1,054 | ±15 |
| 48 | ▼ 11 | SIMBA 1.6闭源Speechify | 1,051 | ±14 |
| 49 | ▼ 3 | Sonic English (Oct 2024)闭源Cartesia | 1,050 | ±12 |
| 50 | ▲ 1 | Gemini 2.5 Pro (Dec 2025)闭源Google | 1,043 | ±15 |
智力进步
涨得最多 · Anthropic +57.2目前最高 · Anthropic 64.6最高与最低相差 18.1 分时间范围 2025年11月 – 2026年11月
64.6Anthropic60.0Google56.9OpenAI50.1Meta46.5Kimi