Provider Voice Arena LeaderboardArtificial Analysis
Compare Text to Speech (TTS) models using each provider's own native voices.
Range | Model | Elo | Samples | Arena Voices | Released | API Pricing 1 | |
|---|---|---|---|---|---|---|---|
| 1 | 1-2 |  Eleven v4 Turbo | 1335±1795% interval 1318 to 1352 | 2,103 | Sept 2026 | $40.0 /1M chars | |
| 2 | 1-2 | Eleven v4 | 1330±1695% interval 1314 to 1346 | 2,395 | Sept 2026 | $80.0 /1M chars | |
| 3 | 3 |  Qwen-Audio-3.1-TTS-Plus | 1306±1595% interval 1291 to 1321 | 2,048 | Sept 2026 | $19.3 /1M chars | |
| 4 | 4-5 |  Sonic 3.6 | 1289±1495% interval 1275 to 1303 | 2,861 | Aug 2026 | $49.0 /1M chars | |
| 5 | 4-7 |  Gemini 3.8 Flash TTS | 1282±1495% interval 1268 to 1296 | 3,112 | Sept 2026 | $16.5 /1M chars | |
| 6 | 5-7 |  Realtime TTS-2 | 1269±1495% interval 1255 to 1283 | 2,284 | Aug 2026 | $20.8 /1M chars | |
| 7 | 5-7 | Qwen-Audio-3.0-TTS-Plus | 1268±1495% interval 1254 to 1282 | 2,528 | Sept 2026 | $19.3 /1M chars | |
| 8 | 8-10 | Gemini 3.8 Flash-Lite TTS | 1254±1395% interval 1241 to 1267 | 3,102 | Sept 2026 | $11.0 /1M chars | |
| 9 | 8-10 |  Simba 3.2 | 1246±1295% interval 1234 to 1258 | 3,473 | Jul 2026 | $6.6 /1M chars | |
| 10 | 8-10 |  Fish Audio Drama 3 Preview  | 1246±1495% interval 1232 to 1260 | 2,256 | Oct 2026 | $15.0 /1M chars | |
| 11 | 11-13 |  Breeze TTS 2 | 1230±1395% interval 1217 to 1243 | 2,407 | Aug 2026 | $34.0 /1M chars | |
| 12 | 11-17 |  Voxtral 3 TTS | 1221±1595% interval 1206 to 1236 | 1,749 | Oct 2026 | $13.0 /1M chars | |
| 13 | 11-17 | Realtime TTS-2 Flash | 1218±1395% interval 1205 to 1231 | 2,724 | Aug 2026 | $10.4 /1M chars | |
| 14 | 12-17 | v3 Conversational | 1215±1395% interval 1202 to 1228 | 2,867 | Aug 2026 | $50.0 /1M chars | |
| 15 | 12-17 |  StepAudio 2.5 TTS (Aug 2026) | 1214±1395% interval 1201 to 1227 | 2,304 | Aug 2026 | $85.0 /1M chars | |
| 16 | 12-17 | Gemini 3.1 Flash TTS | 1212±1195% interval 1201 to 1223 | 4,362 | Apr 2026 | $18.3 /1M chars | |
| 17 | 13-17 |  Luna TTS | 1207±1295% interval 1195 to 1219 | 3,635 | Jun 2026 | $15.0 /1M chars | |
| 18 | 18-21 | Sonic 3.5 | 1192±1195% interval 1181 to 1203 | 3,911 | May 2026 | $49.0 /1M chars | |
| 19 | 18-23 |  Ming-3.0-tts | 1189±1495% interval 1175 to 1203 | 2,212 | Oct 2026 | N/A | |
| 20 | 18-24 | Fish Audio S2.1 Pro (Oct 2026) | 1185±1495% interval 1171 to 1199 | 2,165 | Oct 2026 | $15.0 /1M chars | |
| 21 | 18-24 | Eleven v3 | 1184±1095% interval 1174 to 1194 | 5,178 | Feb 2026 | $100.0 /1M chars | |
| 22 | 19-24 |  Soniox TTS Real-Time v2 | 1180±1195% interval 1169 to 1191 | 5,473 | Aug 2026 | $14.2 /1M chars | |
| 23 | 19-25 |  Gradium TTS (Sep 2026) | 1178±1395% interval 1165 to 1191 | 2,289 | Sept 2026 | $47.8 /1M chars | |
| 24 | 21-25 |  Speech 2.8 HD | 1175±1095% interval 1165 to 1185 | 5,287 | Feb 2026 | $100.0 /1M chars | |
| 25 | 22-26 |  Lightning V3.1 Pro (Jul 2026) | 1168±1295% interval 1156 to 1180 | 2,890 | Jul 2026 | $19.5 /1M chars | |
| 26 | 25-29 |  Falcon 2 | 1157±1395% interval 1144 to 1170 | 2,535 | Jul 2026 | $10.0 /1M chars | |
| 27 | 26-29 | Gradium TTS (Aug 2026) | 1155±1295% interval 1143 to 1167 | 2,530 | Aug 2026 | $47.2 /1M chars | |
| 28 | 26-29 | Speech 2.8 Turbo | 1154±1095% interval 1144 to 1164 | 5,049 | Feb 2026 | $60.0 /1M chars | |
| 29 | 26-32 |  SpaceXAI TTS | 1151±1395% interval 1138 to 1164 | 1,969 | Mar 2026 | $15.0 /1M chars | |
| 30 | 29-33 |  Azure HD 2.5 | 1140±1195% interval 1129 to 1151 | 3,054 | Mar 2026 | $22.0 /1M chars | |
| 31 | 29-33 |  Async Flash v1.5 | 1139±1295% interval 1127 to 1151 | 2,750 | May 2026 | $10.1 /1M chars | |
| 32 | 30-34 | Step TTS 2 (Mar 2026) | 1138±1295% interval 1126 to 1150 | 2,859 | Feb 2026 | $40.0 /1M chars | |
| 33 | 30-36 | Speech 2.6 HD | 1130±1195% interval 1119 to 1141 | 4,346 | Oct 2025 | $100.0 /1M chars | |
| 34 | 32-36 | Async Pro v1.0 | 1127±1295% interval 1115 to 1139 | 2,628 | May 2026 | $20.2 /1M chars | |
| 35 | 33-37 | Speech 2.6 Turbo | 1122±1195% interval 1111 to 1133 | 4,618 | Oct 2025 | $60.0 /1M chars | |
| 36 | 33-39 | Simba 3.0 | 1120±1195% interval 1109 to 1131 | 3,077 | Feb 2026 | $6.6 /1M chars | |
| 37 | 35-41 |  Chatterbox HD | 1111±1295% interval 1099 to 1123 | 2,706 | — | May 2025 | $40.0 /1M chars |
| 38 | 36-41 | Speech-02-HD | 1110±1195% interval 1099 to 1121 | 4,687 | Mar 2025 | $100.0 /1M chars | |
| 39 | 36-41 | Fish Audio S2 Pro | 1109±1295% interval 1097 to 1121 | 2,764 | Mar 2026 | $15.0 /1M chars | |
| 40 | 37-44 |  TTS-1 HD | 1104±1195% interval 1093 to 1115 | 3,800 | Nov 2023 | $30.0 /1M chars | |
| 41 | 37-48 | Step Audio EditX (Mar 2026) | 1100±1295% interval 1088 to 1112 | 2,380 | Feb 2026 | N/A | |
| 42 | 40-49 |  Speech 2.6 | 1095±1295% interval 1083 to 1107 | 2,526 | Jul 2026 | N/A | |
| 43 | 40-49 | Turbo v2.5 | 1095±1095% interval 1085 to 1105 | 7,077 | Jul 2024 | $50.0 /1M chars | |
| 44 | 40-49 | Multilingual v2 | 1095±1095% interval 1085 to 1105 | 7,542 | Aug 2023 | $100.0 /1M chars | |
| 45 | 41-49 | Voxtral TTS | 1090±1295% interval 1078 to 1102 | 2,475 | Mar 2026 | $16.0 /1M chars | |
| 46 | 41-49 | Speech-02-Turbo | 1089±1195% interval 1078 to 1100 | 4,026 | Mar 2025 | $60.0 /1M chars | |
| 47 | 42-49 | OpenAudio S1 | 1088±1195% interval 1077 to 1099 | 4,670 | Jun 2025 | $15.0 /1M chars | |
| 48 | 42-49 | TTS-1 | 1088±1095% interval 1078 to 1098 | 6,329 | Nov 2023 | $15.0 /1M chars | |
| 49 | 42-50 | Gemini 2.5 Flash Lite TTS | 1087±1195% interval 1076 to 1098 | 3,344 | Dec 2025 | $9.2 /1M chars | |
| 50 | 49-55 | Flash v2.5 | 1077±1195% interval 1066 to 1088 | 5,863 | Dec 2024 | $50.0 /1M chars | |
| 51 | 50-59 | GPT-Realtime-2 | 1075±1295% interval 1063 to 1087 | 2,343 | May 2026 | $191.6 /1M chars | |
| 52 | 50-59 | Journey | 1074±1295% interval 1062 to 1086 | 2,386 | Dec 2023 | $160.0 /1M chars | |
| 53 | 50-59 | Studio | 1073±1195% interval 1062 to 1084 | 4,618 | Feb 2023 | $160.0 /1M chars | |
| 54 | 50-62 | T2A-01-HD | 1068±1195% interval 1057 to 1079 | 5,103 | Jan 2025 | $50.0 /1M chars | |
| 55 | 50-65 |  Maya 2 Flash | 1067±1395% interval 1054 to 1080 | 2,183 | Jul 2026 | N/A | |
| 56 | 51-65 |  Polly Generative | 1065±1195% interval 1054 to 1076 | 3,857 | May 2024 | $30.0 /1M chars | |
| 57 | 52-65 |  Kokoro 82M v1.0 | 1063±1195% interval 1052 to 1074 | 5,461 | Jan 2025 | $0.7 /1M chars | |
| 58 | 51-65 | Gemini 2.5 Flash TTS (Dec 2025) | 1063±1295% interval 1051 to 1075 | 2,977 | Dec 2025 | $9.2 /1M chars | |
| 59 | 51-65 |  Magpie-Multilingual 357M (Feb 2026) | 1063±1295% interval 1051 to 1075 | 2,325 | Feb 2026 | N/A | |
| 60 | 53-65 | Chirp 3: HD | 1062±1295% interval 1050 to 1074 | 3,470 | Mar 2025 | $30.0 /1M chars | |
| 61 | 54-68 |  Octave 2 | 1058±1295% interval 1046 to 1070 | 3,227 | Oct 2025 | $87.5 /1M chars | |
| 62 | 54-69 |  Flux TTS | 1057±1395% interval 1044 to 1070 | 2,236 | Aug 2026 | $45.0 /1M chars | |
| 63 | 54-68 | Sonic 3 | 1057±1295% interval 1045 to 1069 | 3,589 | Oct 2025 | $49.0 /1M chars | |
| 64 | 54-68 |  MiMo-V2.5-TTS | 1057±1295% interval 1045 to 1069 | 2,151 | - | N/A | |
| 65 | 55-70 |  Coda | 1055±1295% interval 1043 to 1067 | 2,501 | May 2026 | $50.0 /1M chars | |
| 66 | 61-72 | Maya 2 Global | 1049±1395% interval 1036 to 1062 | 2,033 | Jul 2026 | N/A | |
| 67 | 61-72 | Async Flash v1.0 | 1049±1195% interval 1038 to 1060 | 4,116 | Jul 2025 | $10.1 /1M chars | |
| 68 | 61-72 | Maya1 | 1047±1295% interval 1035 to 1059 | 3,344 | Jun 2025 | N/A | |
| 69 | 62-74 |  Higgs Audio V3 TTS | 1045±1395% interval 1032 to 1058 | 2,249 | Jun 2026 | $15.0 /1M chars | |
| 70 | 57-78 | OpenAudio S1 Mini | 1044±2095% interval 1024 to 1064 | 1,682 | Jun 2025 | $15.0 /1M chars | |
| 71 | 66-77 | Gemini 2.5 Pro (Dec 2025) | 1038±1295% interval 1026 to 1050 | 2,499 | Dec 2025 | $18.3 /1M chars | |
| 72 | 66-77 | Octave TTS | 1038±1195% interval 1027 to 1049 | 4,747 | Feb 2025 | $87.5 /1M chars | |
| 73 | 69-79 |  Bland Speech v3 | 1034±1295% interval 1022 to 1046 | 2,552 | Aug 2026 | $40.0 /1M chars | |
| 74 | 66-79 | Azure Neural | 1033±2095% interval 1013 to 1053 | 1,156 | Sept 2018 | $15.0 /1M chars | |
| 75 | 71-79 | Polly Long-Form | 1030±1395% interval 1017 to 1043 | 1,992 | Nov 2023 | $100.0 /1M chars | |
| 76 | 71-79 | T2A-01-Turbo | 1029±1195% interval 1018 to 1040 | 5,012 | Jan 2025 | $30.0 /1M chars | |
| 77 | 71-79 | MAI-Voice-1 | 1028±1295% interval 1016 to 1040 | 2,371 | Apr 2026 | N/A | |
| 78 | 73-79 | Chatterbox  | 1026±1195% interval 1015 to 1037 | 4,735 | May 2025 | $25.0 /1M chars | |
| 79 | 73-79 | Lightning v3.1 | 1023±1295% interval 1011 to 1035 | 2,365 | Mar 2026 | $25.0 /1M chars | |
| 80 | 80-81 |  Raon SpeechLM | 1005±1395% interval 992 to 1018 | 2,223 | Apr 2026 | N/A | |
| 81 | 81 |  Zonos-v0.1 | 1000AnchorFixed at 1000 as the reference point for this board's Elo scale. | 4,884 | Feb 2025 | $20.0 /1M chars | |
| 82 | 82 | Murf Speech Gen 2 | 986±1295% interval 974 to 998 | 5,038 | Jul 2024 | $100.0 /1M chars | |
| 83 | 83-84 | VibeVoice 1.5B | 956±1495% interval 942 to 970 | 2,165 | Aug 2025 | N/A | |
| 84 | 83-85 |  OpenVoice v2 | 954±1395% interval 941 to 967 | 3,086 | — | Apr 2024 | $8.3 /1M chars |
| 85 | 84-87 |  Neuphonic TTS | 942±1495% interval 928 to 956 | 2,496 | Oct 2024 | $20.8 /1M chars | |
| 86 | 84-87 | Qwen3 TTS Flash | 941±1495% interval 927 to 955 | 2,223 | Sept 2025 | $11.0 /1M chars | |
| 87 | 85-89 | Qwen3 TTS | 930±1395% interval 917 to 943 | 3,126 | Jan 2026 | $13.8 /1M chars | |
| 88 | 87-90 |  Kugel 3 | 925±1395% interval 912 to 938 | 4,403 | Jun 2026 | N/A | |
| 89 | 87-90 | WaveNet | 918±1295% interval 906 to 930 | 6,404 | Sept 2016 | $4.0 /1M chars | |
| 90 | 88-91 |  XTTS v2 | 914±1595% interval 899 to 929 | 2,708 | — | Nov 2023 | $40.4 /1M chars |
| 91 | 89-94 | Mist V2 | 904±1595% interval 889 to 919 | 3,054 | Mar 2025 | $31.1 /1M chars | |
| 92 | 91-95 | Polly Neural | 895±1495% interval 881 to 909 | 2,977 | Jul 2019 | $16.0 /1M chars | |
| 93 | 91-95 | Neural2 | 894±1295% interval 882 to 906 | 6,689 | Jun 2022 | $16.0 /1M chars | |
| 94 | 91-95 |  StyleTTS 2 | 890±1595% interval 875 to 905 | 2,677 | — | Jun 2023 | $2.8 /1M chars |
| 95 | 92-95 | Standard | 887±1295% interval 875 to 899 | 6,769 | Aug 2018 | $4.0 /1M chars | |
| 96 | 95-96 |  Noiz TTS | 873±1595% interval 858 to 888 | 2,434 | Nov 2025 | N/A | |
| 97 | 97-98 |  MetaVoice v1 | 846±1895% interval 828 to 864 | 1,655 | — | Feb 2024 | N/A |
| 98 | 97-98 | Polly Standard | 830±1695% interval 814 to 846 | 3,589 | Nov 2016 | $4.0 /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.
Frequently Asked Questions
The top Text to Speech models by Elo rating are: 1. Eleven v4 Turbo (Elo 1335), 2. Eleven v4 (Elo 1330), 3. Qwen-Audio-3.1-TTS-Plus (Elo 1306), 4. Sonic 3.6 (Elo 1289), and 5. Gemini 3.8 Flash TTS (Elo 1282). Rankings are based on blind user votes in the Speech Arena.
Kokoro 82M v1.0 is the most affordable at $0.65 per 1M characters with an Elo score of 1063. Other affordable options include StyleTTS 2 at $2.82 per 1M characters.
The top 5 open weights Text to Speech models are: 1. Fish Audio Drama 3 Preview (Elo 1246), 2. Breeze TTS 2 (Elo 1230), 3. Ming-3.0-tts (Elo 1189), 4. Fish Audio S2 Pro (Elo 1109), and 5. Step Audio EditX (Mar 2026) (Elo 1100).