Gradium's On-Device Phonon TTS Beats Models 100x Its Size on Key Benchmarks
LysandreJik · x · 2026-11-05
A small on-device TTS model is competing with models up to 100× its size: @GradiumAI's new Phonon model takes first place on the Open TTS Leaderboard for intelligibility in German and Spanish, top-three intelligibility in every language evaluated, and leading CPU streaming time-to-first-audio.
The results also show why TTS models shouldn't be judged on a single overall ranking. The Open TTS Leaderboard separates dimensions: intelligibility (via word/character error rates), throughput and time-to-first-audio, predicted naturalness, and speaker similarity for voice cloning. Leading on one language's intelligibility or CPU latency is a specific result, not proof of overall superiority.
Related event: Gradium's tiny open-source Phonon TTS beats models 100x larger(3 posts)→
More from Models
- FrontierPhysics benchmark: Claude Opus 5.5 tops end-to-end physics research at 27.4% — xeophon · 2026-11-06
- Why AI writing feels annoying despite models winning literary competitions — cis_female · 2026-11-06
- METR chart reveals unreleased Anthropic Model 4 scoring 178.6 on AECI — ResultBackground2450 · 2026-11-06
- Marin's open 535B-A23B MoE finishes main pretraining on 18T tokens, checkpoints soon — dlwh · 2026-11-06
- Artificial Analysis launches Intelligence Index v5 and new Progress Index, reveals eval roadmap — ArtificialAnlys · 2026-11-06
- Fastino open-sources GLiDE: five decision models (0.4B-27B) claiming a new Pareto frontier — HowDevelop · 2026-11-06