模型排行

排名	变化	模型	得分	置信区间
1	—	Sonic 3.6闭源Cartesia	1,282	±17
2	—	Realtime TTS-2闭源Inworld	1,252	±18
3	—	Qwen-Audio-3.0-TTS-Plus闭源Alibaba	1,241	±14
4	—	Simba 3.2闭源SpeechifyAI	1,240	±14
5	—	Luna TTS闭源VUI Labs	1,228	±14
6	—	Realtime TTS-2 Flash - Research Preview闭源Inworld	1,222	±15
7	—	v3 Conversational闭源ElevenLabs	1,210	±15
8	—	Gemini 3.1 Flash TTS闭源Google	1,208	±12
9	—	StepAudio 2.5 TTS (Aug 2026)闭源StepFun	1,205	±17
10	—	Sonic 3.5闭源Cartesia	1,193	±12
11	—	Lightning V3.1 Pro (Jul 2026)闭源Smallest.ai	1,190	±14
12	—	Soniox TTS Real-Time v2闭源Soniox	1,179	±11
13	—	Eleven v3闭源ElevenLabs	1,175	±11
14	—	Speech 2.8 HD闭源MiniMax	1,171	±11
15	—	Falcon 2闭源Murf AI	1,158	±15
16	—	Speech 2.8 Turbo闭源MiniMax	1,152	±11
17	—	Gradium TTS (Aug 2026)闭源Gradium	1,149	±15
18	—	Async Flash v1.5闭源Async	1,141	±13
19	—	Fish Audio S2.1 Pro闭源Fish Audio	1,141	±13
20	—	Step TTS 2 (Mar 2026)闭源StepFun	1,138	±12
21	—	Speech 2.6 HD闭源MiniMax	1,134	±11
22	—	Lightning V3.1 Pro TTS (Jun 2026)闭源Smallest.ai	1,133	±13
23	—	Azure HD 2.5闭源Microsoft	1,131	±12
24	—	Speech 2.6 Turbo闭源MiniMax	1,127	±11
25	—	SpaceXAI TTS闭源SpaceXAI	1,123	±16
26	—	Simba 3.0闭源SpeechifyAI	1,122	±12
27	—	Async Pro v1.0闭源Async	1,121	±13
28	—	Speech-02-HD闭源MiniMax	1,112	±11
29	—	TTS-1 HD闭源OpenAI	1,106	±11
30	—	Turbo v2.5闭源ElevenLabs	1,100	±10
31	—	Multilingual v2闭源ElevenLabs	1,099	±10
32	—	Chatterbox HD闭源Resemble AI	1,093	±12
33	—	TTS-1闭源OpenAI	1,091	±10
34	—	Gemini 2.5 Flash Lite TTS闭源Google	1,084	±12
35	—	Speech 2.6闭源Hithink	1,084	±13
36	—	Speech-02-Turbo闭源MiniMax	1,082	±11
37	—	Studio闭源Google	1,081	±11
38	—	OpenAudio S1闭源Fish Audio	1,080	±11
39	—	Flash v2.5闭源ElevenLabs	1,080	±11
40	—	Journey闭源Google	1,075	±13
41	—	Maya 2 Flash闭源Maya Research	1,072	±14
42	—	Sonic 3闭源Cartesia	1,067	±11
43	—	GPT-Realtime-2闭源OpenAI	1,067	±13
44	—	T2A-01-HD闭源MiniMax	1,066	±11
45	—	Polly Generative闭源Amazon	1,063	±11
46	—	SIMBA 1.6闭源SpeechifyAI	1,062	±12
47	—	Maya 2 Global闭源Maya Research	1,059	±14
47	—	MiMo-V2.5-TTS闭源Xiaomi	1,059	±13
49	—	Coda闭源Rime	1,058	±12
50	—	Chirp 3: HD闭源Google	1,057	±11
51	—	Octave 2闭源Hume AI	1,056	±12
52	—	Gemini 2.5 Flash TTS (Dec 2025)闭源Google	1,056	±12
53	—	OpenAudio S1 Mini闭源Fish Audio	1,050	±19
54	—	Async Flash v1.0闭源Async	1,048	±11
55	—	Polly Long-Form闭源Amazon	1,044	±13
56	—	SIMBA 1.0闭源SpeechifyAI	1,034	±11
57	—	MAI-Voice-1闭源Microsoft	1,034	±12
58	—	Bland Speech v3闭源Bland AI	1,034	±13
59	—	T2A-01-Turbo闭源MiniMax	1,033	±11
60	—	Gemini 2.5 Pro (Dec 2025)闭源Google	1,031	±12
61	—	Octave TTS闭源Hume AI	1,029	±11
62	—	Azure Neural闭源Microsoft	1,029	±19
63	—	MiMo-V2-TTS闭源Xiaomi	1,024	±13
64	—	Lightning v3.1闭源Smallest.ai	1,021	±12
65	—	Raon SpeechLM闭源Krafton	1,009	±13
66	—	Arcana v3闭源Rime	1,005	±13
67	—	Murf Speech Gen 2闭源Murf AI	981	±11
68	—	VibeVoice 1.5B闭源Microsoft	966	±13
69	—	Magpie Multilingual闭源NVIDIA	946	±14
70	—	Qwen3 TTS Flash闭源Alibaba	945	±14
71	—	Neuphonic TTS闭源Neuphonic	937	±13
72	—	Kugel 3闭源KugelAudio	926	±13
73	—	Qwen3 TTS闭源Alibaba	926	±13
74	—	WaveNet闭源Google	915	±12
75	—	Mist V2闭源Rime	901	±14
76	—	Neural2闭源Google	891	±12
77	—	Polly Neural闭源Amazon	889	±14
78	—	Standard闭源Google	882	±12
79	—	Noiz TTS闭源Noiz	869	±15
80	—	Polly Standard闭源Amazon	825	±15

其余 30 个模型

智力进步

每家厂商当时最强的模型在 Artificial Analysis 智力指数上的分数，按模型发布的月份画。

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