Nari Labs launches 50ms TTS endpoint, 10x cheaper than ElevenLabs on Qwen3-TTS
iamaliveix · x · 2026-09-11
Nari Labs launched a TTS inference service claiming a 50 ms time-to-first-audio, the fastest endpoint around, at $5 per 1M characters. Built on Qwen3-TTS 1.7B with their own inference engine, it's 5x faster than Cartesia and 10x cheaper than ElevenLabs, with expressive voices said to beat industry averages. Free for a limited time; the team frames open source as the future of multimodal AI, starting with speech.
Related event: Nari Labs Open-Sources Qwen3-TTS 1.7B: 50ms Latency at a Tenth the Cost(2 posts)→
More from Infra
- SemiAnalysis: OpenAI reportedly negotiating ~1GW TPU commitment while seeking sites — dylan522p · 2026-10-28
- I tested 11 video upscaling methods for AI video: local, Topaz and cloud compared — lopas99 · 2026-10-28
- Thermo-backed CTO says AI is in its dial-up phase, touts disaggregated compute for an ultrafast 'broadband moment' — beffjezos · 2026-10-28
- DeepSeek-V4.1-Flash Q2 runs at 41 tok/s on one RTX 3090, 90% token match vs API — musicurgy · 2026-10-28
- USearch v3 RC ships after 1,000 days, its libraries now embedded in 550K-star projects — viglovikov · 2026-10-28
- SK Hynix says memory demand will outpace supply through 2030 — Chemical_Hawk_6307 · 2026-10-28