Model rankings

Rank	Change	Model	Score	95% interval
1	—	Eleven v4 TurboClosedElevenLabs	1,332	±17
2	—	Eleven v4ClosedElevenLabs	1,329	±16
3	—	Qwen-Audio-3.1-TTS-PlusClosedAlibaba	1,297	±16
4	—	Sonic 3.6ClosedCartesia	1,291	±14
5	—	Gemini 3.8 Flash TTSClosedGoogle	1,280	±14
6	—	Qwen-Audio-3.0-TTS-PlusClosedAlibaba	1,264	±14
7	—	Realtime TTS-2ClosedInworld	1,262	±15
8	—	Gemini 3.8 Flash-Lite TTSClosedGoogle	1,252	±14
9	—	Fish Audio Drama 3 PreviewOpenFish Audio	1,247	±15
10	—	Simba 3.2ClosedSpeechifyAI	1,245	±12
11	—	Breeze TTS 2OpenBreezeBlue	1,222	±14
12	—	StepAudio 2.5 TTS (Aug 2026)ClosedStepFun	1,217	±14
13	—	Realtime TTS-2 FlashClosedInworld	1,217	±13
14	—	Luna TTSClosedVUI Labs	1,213	±12
15	—	Gemini 3.1 Flash TTSClosedGoogle	1,212	±11
16	—	v3 ConversationalClosedElevenLabs	1,212	±13
17	—	Sonic 3.5ClosedCartesia	1,191	±11
18	—	Ming-3.0-ttsOpenAnt Group	1,188	±14
19	—	Fish Audio S2.1 Pro (Oct 2026)ClosedFish Audio	1,187	±14
20	—	Eleven v3ClosedElevenLabs	1,182	±11
21	—	Soniox TTS Real-Time v2ClosedSoniox	1,179	±11
22	—	Gradium TTS (Sep 2026)ClosedGradium · 1 more variant	1,175	±13
23	—	Speech 2.8 HDClosedMiniMax	1,174	±11
24	—	Lightning V3.1 Pro (Jul 2026)ClosedSmallest.ai	1,168	±12
25	—	Speech 2.8 TurboClosedMiniMax	1,155	±11
26	—	Falcon 2ClosedMurf AI	1,153	±13
27	—	SpaceXAI TTSClosedSpaceXAI	1,149	±13
28	—	Azure HD 2.5ClosedMicrosoft	1,140	±12
29	—	Async Flash v1.5ClosedAsync	1,139	±12
30	—	Step TTS 2 (Mar 2026)ClosedStepFun	1,135	±12
31	—	Speech 2.6 HDClosedMiniMax	1,129	±11
32	—	Async Pro v1.0ClosedAsync	1,127	±12
33	—	Simba 3.0ClosedSpeechifyAI	1,121	±11
34	—	Speech 2.6 TurboClosedMiniMax	1,121	±11
35	—	Fish Audio S2 ProOpenFish Audio	1,113	±12
36	—	Speech-02-HDClosedMiniMax	1,111	±11
37	—	Chatterbox HDClosedResemble AI	1,109	±12
38	—	TTS-1 HDClosedOpenAI	1,102	±11
39	—	Step Audio EditX (Mar 2026)OpenStepFun	1,099	±12
40	—	Turbo v2.5ClosedElevenLabs	1,098	±10
41	—	Speech 2.6ClosedHithink	1,096	±12
42	—	Multilingual v2ClosedElevenLabs	1,096	±10
43	—	Speech-02-TurboClosedMiniMax	1,091	±11
44	—	Voxtral TTSOpenMistral	1,091	±12
45	—	TTS-1ClosedOpenAI	1,089	±10
46	—	Gemini 2.5 Flash Lite TTSClosedGoogle	1,086	±11
47	—	OpenAudio S1ClosedFish Audio	1,085	±11
48	—	StudioClosedGoogle	1,077	±11
49	—	Flash v2.5ClosedElevenLabs	1,077	±11
50	—	JourneyClosedGoogle	1,077	±12
51	—	GPT-Realtime-2ClosedOpenAI	1,076	±12
52	—	T2A-01-HDClosedMiniMax	1,067	±11
53	—	Polly GenerativeClosedAmazon	1,066	±11
54	—	Maya 2 FlashClosedMaya Research	1,065	±13
55	—	Kokoro 82M v1.0OpenKokoro	1,065	±11
56	—	Magpie-Multilingual 357M (Feb 2026)OpenNVIDIA	1,063	±12
57	—	Gemini 2.5 Flash TTS (Dec 2025)ClosedGoogle	1,062	±12
58	—	Chirp 3: HDClosedGoogle	1,061	±12
59	—	Sonic 3ClosedCartesia	1,056	±12
60	—	CodaClosedRime	1,056	±12
61	—	Octave 2ClosedHume AI	1,055	±12
62	—	MiMo-V2.5-TTSClosedXiaomi	1,052	±12
63	—	Flux TTSClosedDeepgram	1,051	±13
64	—	Async Flash v1.0ClosedAsync	1,050	±11
65	—	Maya 2 GlobalClosedMaya Research	1,048	±13
66	—	Maya1OpenMaya Research	1,047	±12
67	—	OpenAudio S1 MiniOpenFish Audio	1,044	±20
68	—	Higgs Audio V3 TTSOpenBoson AI	1,044	±13
69	—	Octave TTSClosedHume AI	1,038	±11
70	—	Gemini 2.5 Pro (Dec 2025)ClosedGoogle	1,038	±12
71	—	Bland Speech v3ClosedBland AI	1,037	±13
72	—	Azure NeuralClosedMicrosoft	1,034	±20
73	—	T2A-01-TurboClosedMiniMax	1,030	±11
74	—	MAI-Voice-1ClosedMicrosoft	1,030	±12
75	—	Polly Long-FormClosedAmazon	1,029	±13
76	—	ChatterboxOpenResemble AI	1,024	±11
77	—	Lightning v3.1ClosedSmallest.ai	1,022	±12
78	—	Raon SpeechLMClosedKrafton	1,005	±13
79	—	Zonos-v0.1OpenZyphra	1,000	±0
80	—	Murf Speech Gen 2ClosedMurf AI	986	±12
81	—	VibeVoice 1.5BOpenMicrosoft	956	±14
82	—	OpenVoice v2OpenOpenVoice	956	±13
83	—	Neuphonic TTSClosedNeuphonic	943	±14
84	—	Qwen3 TTS FlashClosedAlibaba	942	±14
85	—	Qwen3 TTSClosedAlibaba	930	±13
86	—	Kugel 3ClosedKugelAudio	930	±13
87	—	WaveNetClosedGoogle	918	±12
88	—	XTTS v2OpenCoqui	916	±15
89	—	Mist V2ClosedRime	904	±15
90	—	Polly NeuralClosedAmazon	897	±14
91	—	Neural2ClosedGoogle	894	±12
92	—	StyleTTS 2OpenStyleTTS	892	±15
93	—	StandardClosedGoogle	886	±13
94	—	Noiz TTSClosedNoiz	875	±15
95	—	MetaVoice v1OpenMetaVoice	847	±18
96	—	Polly StandardClosedAmazon	831	±16

46 more models

Intelligence over time

Each maker's best model at the time on the Artificial Analysis Intelligence Index, by the month it was released.

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