Gradium-TTS tops voice latency board at 68 ms to first audio
mattturck · x · 2026-10-07
GradiumAI's Gradium-TTS now ranks #1 on Voice Arena's US English latency board with a median 68 ms time to first audio — the fastest model measured to date.
- Ahead of Speechify's Simba 3.2 (123 ms) and Inworld TTS 2 (168.5 ms); over 3x faster than its own previous 236 ms
- Quality holds up: 1007 Elo (#10 of 21) from blind native-speaker votes, placing it on the speed-quality frontier alongside Simba 3.2, Inworld TTS 2 and Cartesia's Sonic-3.6
- Methodology counts leading silence and uses fresh text per request, reflecting real listener experience
Related event: Gradium-TTS Tops Voice Arena Latency Chart with 68ms First Audio(2 posts)→
More from Multimodal
- Creator tests Runway's latest tools with cinematic AI car commercial "Speedhunter" — Uncanny_Harry · 2026-10-07
- First-try image comparison: GPT edges out Nano Banana 2.1 on detail — techhalla · 2026-10-07
- Photo round: GPT Image 2.5 edges Nano Banana 2.1 on detail in 20-prompt test — techhalla · 2026-10-07
- Text and still-life round: neither model fumbles text, GPT slightly ahead on aesthetics — techhalla · 2026-10-07
- Nano Banana 2.1 vs GPT Image 2.5: 20-prompt test scored on adherence, aesthetics, composition — techhalla · 2026-10-07
- Six hours after launch: is Nano Banana 2.1 better than GPT Image 2.5? — techhalla · 2026-10-07