Controlled Voice Arena LeaderboardArtificial Analysis

Compare Text to Speech (TTS) models using the same 8 cloned voices (4 US, 4 UK).

Range
Model
Elo
Samples
Released
API Pricing 1
11-2
HeyGen logo
HeyGen Voice
1195±1695% interval 1179 to 1211
1,684Oct 2026$30.0 /1M chars
22
Alibaba logo
Qwen-Audio-3.1-TTS-Plus
1180±1495% interval 1166 to 1194
2,046Sept 2026$19.3 /1M chars
33-4
ElevenLabs logo
Eleven v4 Turbo
1163±1495% interval 1149 to 1177
2,326Sept 2026$40.0 /1M chars
43-4
ElevenLabs logo
Eleven v4
1158±1495% interval 1144 to 1172
2,287Sept 2026$80.0 /1M chars
55-6
Cartesia logo
Sonic 3.6
1143±1495% interval 1129 to 1157
2,900Aug 2026$49.0 /1M chars
65-6
Inworld logo
Realtime TTS-2
1140±1495% interval 1126 to 1154
2,870Aug 2026$20.8 /1M chars
77-8
Alibaba logo
Qwen-Audio-3.0-TTS-Plus
1125±1495% interval 1111 to 1139
2,615Sept 2026$19.3 /1M chars
87-8
Cartesia logo
Sonic 3.5
1116±1295% interval 1104 to 1128
4,886May 2026$49.0 /1M chars
99-10
SpeechifyAI logo
Simba 3.2
1088±1495% interval 1074 to 1102
2,553Jul 2026$6.6 /1M chars
1010-12
StepFun logo
StepAudio 2.5 TTS
1074±1395% interval 1061 to 1087
3,462Apr 2026$85.0 /1M chars
1110-12
ElevenLabs logo
Eleven v3
1072±1295% interval 1060 to 1084
4,086Feb 2026$100.0 /1M chars
1210-12
Inworld logo
Realtime TTS-2 Flash
1072±1395% interval 1059 to 1085
2,902Aug 2026$10.4 /1M chars
1312-18
Google logo
Gemini 3.8 Flash-Lite TTS
1058±1495% interval 1044 to 1072
2,808Sept 2026$11.0 /1M chars
1411-18
Fish Audio logo
Fish Audio Drama 3 PreviewHugging Face
1058±1595% interval 1043 to 1073
1,804Oct 2026$15.0 /1M chars
1513-18
ElevenLabs logo
v3 Conversational
1057±1495% interval 1043 to 1071
2,487Aug 2026$50.0 /1M chars
1613-18
SpaceXAI logo
SpaceXAI TTS
1055±1395% interval 1042 to 1068
3,116Mar 2026$15.0 /1M chars
1713-18
Gradium logo
Gradium TTS (Sep 2026)
1054±1495% interval 1040 to 1068
2,014Sept 2026$47.8 /1M chars
1813-18
Google logo
Gemini 3.8 Flash TTS
1048±1495% interval 1034 to 1062
2,736Sept 2026$16.5 /1M chars
1919-21
SpeechifyAI logo
Simba 3.0
1032±1395% interval 1019 to 1045
2,958Feb 2026$6.6 /1M chars
2019-24
MiniMax logo
Speech 2.8 HD
1019±1395% interval 1006 to 1032
3,536Feb 2026$100.0 /1M chars
2119-24
Gradium logo
Gradium TTS (Aug 2026)
1019±1495% interval 1005 to 1033
2,397Aug 2026$47.2 /1M chars
2220-24
MiniMax logo
Speech 2.8 Turbo
1018±1395% interval 1005 to 1031
3,751Feb 2026$60.0 /1M chars
2320-24
BreezeBlue logo
Breeze TTS 2Hugging Face
1017±1495% interval 1003 to 1031
2,401Aug 2026$34.0 /1M chars
2420-27
Fish Audio logo
Fish Audio S2.1 Pro (Oct 2026)
1015±1595% interval 1000 to 1030
1,751Oct 2026$15.0 /1M chars
2524-29
Fish Audio logo
OpenAudio S1 MiniHugging Face
1002±1495% interval 988 to 1016
2,758Jun 2025$15.0 /1M chars
2624-29
Mistral logo
Voxtral TTSHugging Face
1002±1495% interval 988 to 1016
2,772Mar 2026$16.0 /1M chars
2725-29
Fish Audio logo
Fish Audio S2 ProHugging Face
1001±1395% interval 988 to 1014
4,216Mar 2026$15.0 /1M chars
2828
Fish Audio logo
OpenAudio S1
1000AnchorFixed at 1000 as the reference point for this board's Elo scale.
5,860Jun 2025$15.0 /1M chars
2925-31
ElevenLabs logo
Multilingual v2
997±1395% interval 984 to 1010
3,583Aug 2023$100.0 /1M chars
3029-31
Fish Audio logo
Fish Audio S1
987±1395% interval 974 to 1000
3,567Oct 2025$15.0 /1M chars
3129-31
ElevenLabs logo
Flash v2.5
986±1395% interval 973 to 999
3,530Dec 2024$50.0 /1M chars
3231-34
ElevenLabs logo
Turbo v2.5
973±1495% interval 959 to 987
2,764Jul 2024$50.0 /1M chars
3332-34
KugelAudio logo
Kugel 3
962±1395% interval 949 to 975
4,035Jun 2026N/A
3432-35
Boson AI logo
Higgs Audio V3 TTSHugging Face
960±1495% interval 946 to 974
2,434Jun 2026$15.0 /1M chars
3534-37
Cartesia logo
Sonic 3
947±1495% interval 933 to 961
2,763Oct 2025$49.0 /1M chars
3635-38
Soniox logo
Soniox TTS Real-Time v2
944±1595% interval 929 to 959
2,244Aug 2026$14.2 /1M chars
3735-39
NVIDIA logo
Magpie TTS Zeroshot
939±1795% interval 922 to 956
1,060Jul 2026N/A
3836-40
Resemble AI logo
ChatterboxGitHub
929±1595% interval 914 to 944
2,720May 2025$25.0 /1M chars
3937-40
Krafton logo
Raon SpeechLM
928±1595% interval 913 to 943
2,239Apr 2026N/A
4038-40
Noiz logo
Noiz TTS
918±1595% interval 903 to 933
2,303Nov 2025N/A
4141
Smallest.ai logo
Lightning v3.1
886±1595% interval 871 to 901
2,351Mar 2026$25.0 /1M chars
4242
Coqui logo
XTTS v2Hugging Face
830±1795% interval 813 to 847
2,500Nov 2023$40.4 /1M chars
4343
OpenVoice logo
OpenVoice v2Hugging Face
811±1895% interval 793 to 829
2,190Apr 2024$8.3 /1M chars

1 API Pricing reflects the cost to generate 1M characters on the model creator's API with the model's default settings

2 Open Weights includes models that can be self-hosted using downloadable model weights or a provider-supplied inference runtime. Open Weights Commercial includes models whose published license permits commercial use, including use subject to license conditions (e.g. an NVIDIA AI Enterprise subscription), and excludes models that require a separately negotiated commercial agreement.