모델 순위

순위	변동	모델	점수	95% 구간
1	—	Fun-Realtime-TTS클로즈드Alibaba	1,226	±16
2	—	Gemini 3.1 Flash TTS클로즈드Google	1,215	±13
3	▲ 2	Realtime TTS-2 - Research Preview클로즈드Inworld	1,208	±14
4	—	Sonic 3.5클로즈드Cartesia	1,197	±15
5	▼ 2	xAI Text to Speech클로즈드xAI	1,196	±14
6	—	Realtime TTS 1.5 Max클로즈드Inworld	1,195	±14
7	—	Async Flash v1.5클로즈드Async	1,181	±15
8	—	StepAudio 2.5 TTS클로즈드StepFun	1,177	±16
9	▲ 1	Eleven v3클로즈드ElevenLabs	1,176	±12
10	▼ 1	Speech 2.8 HD클로즈드MiniMax	1,170	±12
11	▲ 1	Inworld TTS 1 Max클로즈드Inworld	1,161	±13
12	▼ 1	Async Pro v1.0클로즈드Async	1,159	±15
13	▲ 2	Step TTS 2 (Mar 2026)클로즈드StepFun	1,155	±14
14	—	Speech 2.8 Turbo클로즈드MiniMax	1,152	±12
15	▼ 2	Realtime TTS 1.5 Mini클로즈드Inworld	1,151	±12
16	—	Speech 2.6 Turbo클로즈드MiniMax	1,137	±11
17	—	Speech 2.6 HD클로즈드MiniMax	1,136	±12
18	—	SIMBA 3.0클로즈드Speechify	1,129	±13
19	▲ 1	Azure HD 2.5클로즈드Microsoft	1,121	±13
20	▲ 1	Inworld TTS 1클로즈드Inworld	1,121	±12
21	▼ 2	Fish Audio S2 Pro오픈Fish Audio	1,117	±14
22	▲ 2	Turbo v2.5클로즈드ElevenLabs	1,109	±10
23	—	Speech-02-HD클로즈드MiniMax	1,108	±12
24	▲ 1	Multilingual v2클로즈드ElevenLabs	1,107	±10
25	▼ 3	Step Audio EditX (Mar 2026)오픈StepFun	1,107	±14
26	—	Gradium TTS클로즈드Gradium	1,098	±15
27	▲ 5	TTS-1 HD클로즈드OpenAI	1,096	±12
28	▼ 1	Flash v2.5클로즈드ElevenLabs	1,095	±11
29	▼ 1	Lightning V3.1 Pro클로즈드Smallest.ai	1,090	±14
30	—	Speech-02-Turbo클로즈드MiniMax	1,090	±12
31	▼ 2	TTS-1클로즈드OpenAI	1,086	±10
32	▼ 1	Chatterbox HD클로즈드Resemble AI	1,086	±13
33	▲ 1	Sonic 3클로즈드Cartesia	1,083	±12
34	▼ 1	Gemini 2.5 Flash Lite TTS클로즈드Google	1,082	±12
35	—	Studio클로즈드Google	1,081	±11
36	▲ 1	MiMo-V2.5-TTS클로즈드Xiaomi	1,078	±14
37	▼ 1	Voxtral TTS오픈Mistral	1,077	±14
38	—	Polly Generative클로즈드Amazon	1,072	±12
39	—	Azure Neural클로즈드Microsoft	1,067	±12
40	▲ 2	Polly Long-Form클로즈드Amazon	1,064	±13
41	▼ 1	T2A-01-HD클로즈드MiniMax	1,064	±11
42	▲ 1	Kokoro 82M v1.0오픈Kokoro	1,061	±11
43	▼ 2	OpenAudio S1클로즈드Fish Audio	1,060	±11
44	▲ 4	SIMBA 1.6클로즈드Speechify	1,060	±13
45	▲ 2	Magpie-Multilingual 357M (Feb 2026)오픈NVIDIA	1,060	±14
46	▼ 1	Octave 2클로즈드Hume AI	1,060	±12
47	▼ 3	Async Flash v1.0클로즈드Async	1,059	±11
48	▼ 2	Chirp 3: HD클로즈드Google	1,056	±12
49	▲ 4	Maya1오픈Maya Research	1,051	±12
50	▲ 1	Journey클로즈드Google	1,050	±13
51	▼ 2	Sonic English (Oct 2024)클로즈드Cartesia	1,048	±12
52	—	Coda클로즈드Rime	1,044	±14
53	▼ 3	Gemini 2.5 Pro (Dec 2025)클로즈드Google	1,037	±12
54	▲ 2	SIMBA 1.0클로즈드Speechify	1,034	±11
55	—	Gemini 2.5 Flash TTS (Dec 2025)클로즈드Google	1,032	±12
56	▼ 2	Lightning v3.1클로즈드Smallest.ai	1,030	±13
57	▲ 1	T2A-01-Turbo클로즈드MiniMax	1,028	±11
58	▼ 1	Octave TTS클로즈드Hume AI	1,027	±11
59	—	MiMo-V2-TTS클로즈드Xiaomi	1,012	±14
60	—	Fish Speech 1.5오픈Fish Audio	1,008	±11
61	▲ 1	Magpie-Multilingual 357M오픈NVIDIA	1,008	±11
62	▼ 1	Arcana v3클로즈드Rime	1,006	±14
63	—	MAI-Voice-1클로즈드Microsoft	1,004	±13
64	▲ 1	Chatterbox오픈Resemble AI	1,004	±11
65	▼ 1	Zonos-v0.1오픈Zyphra	1,000	±0
66	▲ 4	VibeVoice 7B오픈Microsoft	968	±13
67	▲ 1	Murf Speech Gen 2클로즈드Murf AI	967	±11
68	▼ 1	LMNT클로즈드LMNT	967	±12
69	▼ 3	VibeVoice 1.5B클로즈드Microsoft	961	±13
70	▼ 1	OpenVoice v2오픈OpenVoice	960	±13
71	—	Magpie Multilingual클로즈드NVIDIA	939	±14
72	—	Neuphonic TTS클로즈드Neuphonic	939	±13
73	▲ 1	Qwen3 TTS Flash클로즈드Alibaba	930	±14
74	▼ 1	Qwen3 TTS클로즈드Alibaba	917	±13
75	—	XTTS v2오픈Coqui	914	±15
76	▲ 1	WaveNet클로즈드Google	895	±11
77	▲ 2	StyleTTS 2오픈StyleTTS	889	±15
78	—	Polly Neural클로즈드Amazon	881	±14
79	▼ 3	Mist V2클로즈드Rime	881	±15
80	▲ 1	Standard클로즈드Google	876	±11
81	▼ 1	Neural2클로즈드Google	875	±12
82	—	Noiz TTS클로즈드Noiz	853	±17
83	—	MetaVoice v1오픈MetaVoice	838	±18
84	—	Falcon (Beta)클로즈드Murf AI	826	±15
85	—	Polly Standard클로즈드Amazon	812	±15

모델 35개 더 보기

지능의 발전

각 개발사가 그 시점에 가진 최고 모델의 Artificial Analysis Intelligence Index를 모델 공개 월 기준으로 보여 줍니다.

가장 많이 오름 · Anthropic +57.2현재 최고 · Anthropic 64.6최고와 최저의 차이 18.1점기간 2025년 11월 – 2026년 11월

64.6Anthropic60.0Google56.9OpenAI50.1Meta46.5Kimi