Cartesia Sonic 3.6 tops Artificial Analysis speech Elo leaderboard at 1123
ArtificialAnlys · x · 2026-08-18
Artificial Analysis's controlled voice leaderboard (same 8 cloned voices, 4 US / 4 UK) has a new leader: Cartesia Sonic 3.6 at Elo 1123, generating 136.1 chars/sec at $49 per 1M characters, ahead of its predecessor Sonic 3.5 at 1102.
ElevenLabs Eleven v3 (1059, $100/1M chars) takes third, followed by StepFun StepAudio 2.5 TTS, Inworld Realtime TTS 1.5 Max and its Research Preview. MiniMax Speech 2.8 HD and Turbo sit at 1027/1026. Open-weight entries include Mistral Voxtral TTS (1011) and Fish Audio S2 Pro (1004); the cheapest top model is SpeechifyAI Simba 3.0 at $12.5/1M chars (1017).
Related event: Cartesia Releases Sonic-3.6, Topping Speech Leaderboards(4 posts)→
More from Multimodal
- Train Your Own LTX IC-LoRA With Grok Doing the Heavy Lifting — PurzBeats · 2026-10-16
- Hugging Face launches Open TTS Leaderboard: objective metrics to rank 8K+ open TTS models — dl_weekly · 2026-10-16
- Flux 3 Klein excels at lighting and atmosphere but fumbles living subjects — Schattenmensch · 2026-10-16
- Circus clown nails tightrope stunts in impressive Seedance 2.5 video demo — azed_ai · 2026-10-16
- ComfyUI puts up $25,000 to counter Higgsfield's broken open-source promise — PurzBeats · 2026-10-16
- Underwater brand films built from one jellyfish with an open-source Minimax template — w1kke · 2026-10-16