Breeze TTS 2 tops open weights TTS leaderboard on Artificial Analysis
multimodalart · x · 2026-08-26
Breeze TTS 2 is now available on Hugging Face and ranks as the #1 open weights TTS model on the Artificial Analysis Provider Voices Speech Arena.
- Performance: It ranks #1 among open weights models with an Elo of 1,215, surpassing the next best open model, Fish Audio S2 Pro, by 90 points. It ranks #6 overall out of 100+ models.
- Features: Supports 50 languages; includes Voice Design (prompt a voice), Voice Reference (upload a voice), and Voice Direction (upload & prompt a voice), as well as streaming generation.
Related event: Breeze TTS 2 Tops Open-Weight TTS Leaderboard, License Draws Criticism(5 posts)→
More from Models
- Unredacted Anthropic filing exposes internal data-quality debate months before Opus 3 cutoff — zacharynado · 2026-10-15
- Google says teams across the company are using Gemini 4 Argon on its hardest engineering problems — CurieuxExplorer · 2026-10-15
- Perplexity's new embedding models: index with a 9B model, search with 0.6B — antoine_chaffin · 2026-10-15
- 797GB of n-gram counts in CPU RAM let 1B models match 57% more compute — PangWeiKoh · 2026-10-15
- Leak: Moonshot's Kimi flagship 'x4' done with pre-training, K3.1 in internal testing — lyraxana · 2026-10-15
- Unsloth ships GGUF quants for Nemotron 3.5: Bolt-74B runs in 24GB RAM — danielhanchen · 2026-10-15