Nari Labs launches 40ms speech-to-text at $0.06/hr, claims 9x cheaper than Gemini Transcribe
alexcovo_eth · x · 2026-09-12
Nari Labs announced two speech-to-text offerings:
- World's fastest STT endpoint: 40 ms time-to-final-segment (TTFS).
- World's cheapest streaming STT: $0.06/hr, powered by Qwen3-ASR 1.7B on Nari's own inference engine — claimed 3x faster than ElevenLabs Scribe V2 and 9x cheaper than Gemini Transcribe 3.5, with Top 3 accuracy on Coval benchmarks.
- Free for a limited time, framed as part of Nari's mission that "open-source will win — not just in LLMs, but multimodal AI."
More from Infra
- Multi-model on one node: 79GB idle GPU memory reclaimed with sub-200ms warmups — ahstanin · 2026-09-12
- Intel crosses 1 million wafers on High-NA EUV machines, a first; TSMC not adopting until 2030 — Beth_Kindig · 2026-09-12
- Foresight Institute offers up to $100K grants for local AI compute projects — niloofar_mire · 2026-09-12
- Building a $15k Local Inference Rig for DeepSeek V4 Flash: MI210 vs A40? — _TheWolfOfWalmart_ · 2026-09-12
- Cloudflare quietly cuts default Workflows run retention from 30 days to 7 — HaktanSuren · 2026-09-12
- MiniMax H3 ecosystem roundup: multi-GPU recipes, 24GB VRAM runtime, turbo LoRAs — optimisticalish · 2026-09-12