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 | 1332±1995% interval 1313 to 1351 | 1,689 | Sept 2026 | $40.0 /1M chars | |
| 2 | 1-2 | Eleven v4 | 1324±1795% interval 1307 to 1341 | 1,970 | Sept 2026 | $80.0 /1M chars | |
| 3 | 3-4 |  Qwen-Audio-3.1-TTS-Plus | 1295±1795% interval 1278 to 1312 | 1,533 | Sept 2026 | $19.3 /1M chars | |
| 4 | 4-5 |  Sonic 3.6 | 1279±1595% interval 1264 to 1294 | 2,297 | Aug 2026 | $49.0 /1M chars | |
| 5 | 4-6 |  Gemini 3.8 Flash TTS | 1276±1595% interval 1261 to 1291 | 2,526 | Sept 2026 | $16.5 /1M chars | |
| 6 | 5-7 | Qwen-Audio-3.0-TTS-Plus | 1263±1595% interval 1248 to 1278 | 1,923 | Sept 2026 | $19.3 /1M chars | |
| 7 | 6-9 |  Realtime TTS-2 | 1252±1695% interval 1236 to 1268 | 1,693 | Aug 2026 | $20.8 /1M chars | |
| 8 | 7-9 | Gemini 3.8 Flash-Lite TTS | 1243±1595% interval 1228 to 1258 | 2,458 | Sept 2026 | $11.0 /1M chars | |
| 9 | 7-9 |  Simba 3.2 | 1241±1395% interval 1228 to 1254 | 2,826 | Jul 2026 | $6.6 /1M chars | |
| 10 | 10-13 |  Luna TTS | 1223±1395% interval 1210 to 1236 | 2,967 | Jun 2026 | $15.0 /1M chars | |
| 11 | 10-15 |  Breeze TTS 2  | 1219±1595% interval 1204 to 1234 | 1,722 | Aug 2026 | $34.0 /1M chars | |
| 12 | 10-15 | Realtime TTS-2 Flash | 1215±1495% interval 1201 to 1229 | 2,103 | Aug 2026 | $10.4 /1M chars | |
| 13 | 10-15 |  StepAudio 2.5 TTS (Aug 2026) | 1212±1595% interval 1197 to 1227 | 1,645 | Aug 2026 | $85.0 /1M chars | |
| 14 | 11-15 | Gemini 3.1 Flash TTS | 1210±1295% interval 1198 to 1222 | 3,779 | Apr 2026 | $18.3 /1M chars | |
| 15 | 12-15 | v3 Conversational | 1205±1495% interval 1191 to 1219 | 2,192 | Aug 2026 | $50.0 /1M chars | |
| 16 | 16-17 | Sonic 3.5 | 1190±1295% interval 1178 to 1202 | 3,267 | May 2026 | $49.0 /1M chars | |
| 17 | 17-20 |  Soniox TTS Real-Time v2 | 1179±1195% interval 1168 to 1190 | 4,921 | Aug 2026 | $14.2 /1M chars | |
| 18 | 17-20 | Eleven v3 | 1174±1195% interval 1163 to 1185 | 4,617 | Feb 2026 | $100.0 /1M chars | |
| 19 | 17-20 |  Speech 2.8 HD | 1173±1195% interval 1162 to 1184 | 4,678 | Feb 2026 | $100.0 /1M chars | |
| 20 | 17-20 |  Lightning V3.1 Pro (Jul 2026) | 1171±1395% interval 1158 to 1184 | 2,244 | Jul 2026 | $19.5 /1M chars | |
| 21 | 21-24 |  Falcon 2 | 1154±1495% interval 1140 to 1168 | 1,987 | Jul 2026 | $10.0 /1M chars | |
| 22 | 21-23 | Speech 2.8 Turbo | 1153±1195% interval 1142 to 1164 | 4,540 | Feb 2026 | $60.0 /1M chars | |
| 23 | 21-24 |  Gradium TTS (Aug 2026) | 1152±1495% interval 1138 to 1166 | 1,904 | Aug 2026 | $47.2 /1M chars | |
| 24 | 22-28 |  Fish Audio S2.1 Pro | 1140±1395% interval 1127 to 1153 | 2,402 | Jun 2026 | $15.0 /1M chars | |
| 25 | 22-31 |  SpaceXAI TTS | 1138±1595% interval 1123 to 1153 | 1,413 | Mar 2026 | $15.0 /1M chars | |
| 26 | 24-29 | Step TTS 2 (Mar 2026) | 1136±1295% interval 1124 to 1148 | 2,399 | Feb 2026 | $40.0 /1M chars | |
| 27 | 24-31 |  Async Flash v1.5 | 1136±1395% interval 1123 to 1149 | 2,290 | May 2026 | $10.1 /1M chars | |
| 28 | 24-32 |  Azure HD 2.5 | 1131±1295% interval 1119 to 1143 | 2,594 | Mar 2026 | $22.0 /1M chars | |
| 29 | 25-33 | Speech 2.6 HD | 1127±1195% interval 1116 to 1138 | 3,979 | Oct 2025 | $100.0 /1M chars | |
| 30 | 28-33 | Speech 2.6 Turbo | 1124±1195% interval 1113 to 1135 | 4,285 | Oct 2025 | $60.0 /1M chars | |
| 31 | 28-34 | Simba 3.0 | 1124±1295% interval 1112 to 1136 | 2,650 | Feb 2026 | $6.6 /1M chars | |
| 32 | 28-34 | Async Pro v1.0 | 1119±1395% interval 1106 to 1132 | 2,182 | May 2026 | $20.2 /1M chars | |
| 33 | 29-34 | Fish Audio S2 Pro | 1117±1295% interval 1105 to 1129 | 2,403 | Mar 2026 | $15.0 /1M chars | |
| 34 | 31-35 | Speech-02-HD | 1113±1195% interval 1102 to 1124 | 4,382 | Mar 2025 | $100.0 /1M chars | |
| 35 | 34-40 |  TTS-1 HD | 1103±1295% interval 1091 to 1115 | 3,578 | Nov 2023 | $30.0 /1M chars | |
| 36 | 34-42 |  Speech 2.6 | 1102±1395% interval 1089 to 1115 | 2,122 | Jul 2026 | N/A | |
| 37 | 34-42 |  Chatterbox HD | 1101±1395% interval 1088 to 1114 | 2,421 | — | May 2025 | $40.0 /1M chars |
| 38 | 35-42 | Turbo v2.5 | 1097±1095% interval 1087 to 1107 | 6,837 | Jul 2024 | $50.0 /1M chars | |
| 39 | 35-42 | Step Audio EditX (Mar 2026) | 1097±1395% interval 1084 to 1110 | 2,072 | Feb 2026 | N/A | |
| 40 | 35-42 | Multilingual v2 | 1097±1095% interval 1087 to 1107 | 7,275 | Aug 2023 | $100.0 /1M chars | |
| 41 | 36-46 | Speech-02-Turbo | 1091±1295% interval 1079 to 1103 | 3,791 | Mar 2025 | $60.0 /1M chars | |
| 42 | 38-46 | TTS-1 | 1090±1195% interval 1079 to 1101 | 6,115 | Nov 2023 | $15.0 /1M chars | |
| 43 | 41-49 |  Voxtral TTS | 1083±1395% interval 1070 to 1096 | 2,160 | Mar 2026 | $16.0 /1M chars | |
| 44 | 41-49 | Gemini 2.5 Flash Lite TTS | 1082±1295% interval 1070 to 1094 | 3,131 | Dec 2025 | $9.2 /1M chars | |
| 45 | 41-49 | OpenAudio S1 | 1082±1195% interval 1071 to 1093 | 4,481 | Jun 2025 | $15.0 /1M chars | |
| 46 | 42-49 | Studio | 1079±1195% interval 1068 to 1090 | 4,423 | Feb 2023 | $160.0 /1M chars | |
| 47 | 43-49 | Flash v2.5 | 1079±1195% interval 1068 to 1090 | 5,676 | Dec 2024 | $50.0 /1M chars | |
| 48 | 41-53 | GPT-Realtime-2 | 1078±1395% interval 1065 to 1091 | 2,066 | May 2026 | $191.6 /1M chars | |
| 49 | 43-54 | Journey | 1075±1395% interval 1062 to 1088 | 2,217 | Dec 2023 | $160.0 /1M chars | |
| 50 | 46-60 |  Maya 2 Flash | 1067±1495% interval 1053 to 1081 | 1,915 | Jul 2026 | N/A | |
| 51 | 49-59 | T2A-01-HD | 1066±1195% interval 1055 to 1077 | 4,928 | Jan 2025 | $50.0 /1M chars | |
| 52 | 47-60 |  Magpie-Multilingual 357M (Feb 2026) | 1066±1395% interval 1053 to 1079 | 2,078 | Feb 2026 | N/A | |
| 53 | 49-60 |  Polly Generative | 1065±1295% interval 1053 to 1077 | 3,716 | May 2024 | $30.0 /1M chars | |
| 54 | 50-60 |  Kokoro 82M v1.0 | 1064±1195% interval 1053 to 1075 | 5,288 | Jan 2025 | $0.7 /1M chars | |
| 55 | 50-63 | Sonic 3 | 1060±1295% interval 1048 to 1072 | 3,459 | Oct 2025 | $49.0 /1M chars | |
| 56 | 50-63 |  Coda | 1059±1395% interval 1046 to 1072 | 2,258 | May 2026 | $50.0 /1M chars | |
| 57 | 50-63 | Gemini 2.5 Flash TTS (Dec 2025) | 1058±1295% interval 1046 to 1070 | 2,841 | Dec 2025 | $9.2 /1M chars | |
| 58 | 50-63 | Chirp 3: HD | 1056±1295% interval 1044 to 1068 | 3,326 | Mar 2025 | $30.0 /1M chars | |
| 59 | 50-65 |  Flux TTS | 1055±1495% interval 1041 to 1069 | 1,974 | Aug 2026 | $45.0 /1M chars | |
| 60 | 51-65 |  MiMo-V2.5-TTS | 1054±1395% interval 1041 to 1067 | 1,952 | - | N/A | |
| 61 | 55-65 |  Octave 2 | 1051±1295% interval 1039 to 1063 | 3,076 | Oct 2025 | $87.5 /1M chars | |
| 62 | 55-65 | Async Flash v1.0 | 1051±1195% interval 1040 to 1062 | 3,992 | Jul 2025 | $10.1 /1M chars | |
| 63 | 55-67 | Maya1 | 1048±1295% interval 1036 to 1060 | 3,242 | Jun 2025 | N/A | |
| 64 | 54-75 | OpenAudio S1 Mini | 1044±2095% interval 1024 to 1064 | 1,682 | Jun 2025 | $15.0 /1M chars | |
| 65 | 58-73 | Maya 2 Global | 1044±1495% interval 1030 to 1058 | 1,812 | Jul 2026 | N/A | |
| 66 | 63-76 |  Higgs Audio V3 TTS | 1037±1395% interval 1024 to 1050 | 2,064 | Jun 2026 | $15.0 /1M chars | |
| 67 | 64-73 | Octave TTS | 1037±1195% interval 1026 to 1048 | 4,640 | Feb 2025 | $87.5 /1M chars | |
| 68 | 63-76 |  Bland Speech v3 | 1036±1395% interval 1023 to 1049 | 2,352 | Aug 2026 | $40.0 /1M chars | |
| 69 | 60-76 | Azure Neural | 1035±1995% interval 1016 to 1054 | 1,158 | Sept 2018 | $15.0 /1M chars | |
| 70 | 64-76 | Polly Long-Form | 1033±1395% interval 1020 to 1046 | 1,915 | Nov 2023 | $100.0 /1M chars | |
| 71 | 64-76 | Gemini 2.5 Pro (Dec 2025) | 1032±1395% interval 1019 to 1045 | 2,424 | Dec 2025 | $18.3 /1M chars | |
| 72 | 64-76 | MAI-Voice-1 | 1032±1395% interval 1019 to 1045 | 2,221 | Apr 2026 | N/A | |
| 73 | 66-76 | T2A-01-Turbo | 1030±1195% interval 1019 to 1041 | 4,931 | Jan 2025 | $30.0 /1M chars | |
| 74 | 68-76 | Chatterbox  | 1025±1195% interval 1014 to 1036 | 4,634 | May 2025 | $25.0 /1M chars | |
| 75 | 66-76 | Lightning v3.1 | 1024±1395% interval 1011 to 1037 | 2,185 | Mar 2026 | $25.0 /1M chars | |
| 76 | 66-76 | MiMo-V2-TTS | 1024±1495% interval 1010 to 1038 | 1,772 | Mar 2026 | N/A | |
| 77 | 77-78 |  Raon SpeechLM | 1004±1495% interval 990 to 1018 | 2,082 | Apr 2026 | N/A | |
| 78 | 78 |  Zonos-v0.1 | 1000AnchorFixed at 1000 as the reference point for this board's Elo scale. | 4,828 | Feb 2025 | $20.0 /1M chars | |
| 79 | 79 | Murf Speech Gen 2 | 984±1295% interval 972 to 996 | 4,980 | Jul 2024 | $100.0 /1M chars | |
| 80 | 80-81 | VibeVoice 1.5B | 958±1495% interval 944 to 972 | 2,111 | Aug 2025 | N/A | |
| 81 | 80-82 |  OpenVoice v2 | 954±1395% interval 941 to 967 | 2,990 | Apr 2024 | $8.3 /1M chars | |
| 82 | 81-85 | Qwen3 TTS Flash | 943±1495% interval 929 to 957 | 2,136 | Sept 2025 | $11.0 /1M chars | |
| 83 | 82-85 |  Neuphonic TTS | 940±1495% interval 926 to 954 | 2,388 | Oct 2024 | $20.8 /1M chars | |
| 84 | 82-85 |  Kugel 3 | 937±1395% interval 924 to 950 | 4,187 | Jun 2026 | N/A | |
| 85 | 82-85 | Qwen3 TTS | 932±1395% interval 919 to 945 | 2,914 | Jan 2026 | $13.8 /1M chars | |
| 86 | 86-88 | WaveNet | 917±1295% interval 905 to 929 | 6,240 | Sept 2016 | $4.0 /1M chars | |
| 87 | 86-88 |  XTTS v2 | 916±1595% interval 901 to 931 | 2,547 | — | Nov 2023 | $40.4 /1M chars |
| 88 | 86-91 | Mist V2 | 908±1595% interval 893 to 923 | 2,848 | Mar 2025 | $31.1 /1M chars | |
| 89 | 88-93 |  StyleTTS 2 | 895±1595% interval 880 to 910 | 2,453 | — | Jun 2023 | $2.8 /1M chars |
| 90 | 89-93 | Neural2 | 894±1295% interval 882 to 906 | 6,462 | Jun 2022 | $16.0 /1M chars | |
| 91 | 89-93 | Polly Neural | 894±1495% interval 880 to 908 | 2,755 | Jul 2019 | $16.0 /1M chars | |
| 92 | 89-93 | Standard | 885±1295% interval 873 to 897 | 6,508 | Aug 2018 | $4.0 /1M chars | |
| 93 | 89-93 |  Noiz TTS | 883±1595% interval 868 to 898 | 2,018 | Nov 2025 | N/A | |
| 94 | 94 |  MetaVoice v1 | 850±1895% interval 832 to 868 | 1,365 | — | Feb 2024 | N/A |
| 95 | 95 | Polly Standard | 830±1695% interval 814 to 846 | 3,306 | 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 1332), 2. Eleven v4 (Elo 1324), 3. Qwen-Audio-3.1-TTS-Plus (Elo 1295), 4. Sonic 3.6 (Elo 1279), and 5. Gemini 3.8 Flash TTS (Elo 1276). 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 1064. Other affordable options include StyleTTS 2 at $2.82 per 1M characters.
The top 5 open weights Text to Speech models are: 1. Breeze TTS 2 (Elo 1219), 2. Fish Audio S2 Pro (Elo 1117), 3. Step Audio EditX (Mar 2026) (Elo 1097), 4. Voxtral TTS (Elo 1083), and 5. Magpie-Multilingual 357M (Feb 2026) (Elo 1066).