Back Tenstorrent and Smallest.ai have partnered to deliver a production-grade, on-prem voice AI stack.
Tenstorrent, an AI compute company, and Smallest.ai, an AI research lab building real-time voice AI infrastructure, have partnered to deliver a production-grade, on-prem voice AI stack designed to reduce deployment costs while delivering the performance required for real-time voice applications. The joint solution runs Smallest.ai's Lightning V2 real-time text-to-speech model natively on Tenstorrent Galaxy™ Blackhole™ servers, unlocking enterprise-scale voice AI.
Deploying voice AI at scale has historically required companies to make tradeoffs between quality, cost, or control. Running Smallest.ai's Lightning V2 on Tenstorrent’s Blackhole architecture enables companies to run real-time voice agents entirely on-prem with full data sovereignty.
“There shouldn’t need to be a choice between performance and cost – Tenstorrent has built efficient and scalable compute that reduces the cost of existing workflows and makes entirely new workloads practical,” said Grant Jensen, Head of Software Partner Ecosystem at Tenstorrent.
“Tenstorrent’s architecture is fundamentally different from existing paradigms. Working with the NoC and larger SRAM, we unlocked 4x gains at a fraction of the cost of comparable GPU infrastructure, driving a structural shift in inference economics,” said Ranjith, Senior AI Inference Performance Engineer at Smallest.ai.
The joint solution supports high-volume, data-sensitive use cases across financial services, telecom, healthcare, and enterprise environments, enabling low-latency and highly concurrent workloads while giving organizations greater control over sensitive data and sovereign AI infrastructure.
Lightning V2 is available on Tenstorrent hardware today, with usage-based pricing and no upfront commitments. About Tenstorrent
Tenstorrent is an AI compute company led by CEO Jim Keller - architect of Apple A4/A5, AMD Zen, and Tesla's Full Self-Driving chip. The company builds RISC-V-based AI processors and systems for developers, enterprises, and sovereign infrastructure worldwide. In addition to servers and workstations, Tenstorrent licenses its TT-Ascalon RISC-V CPU and Tensix AI cores to chip designers including Samsung and LG. Backed by Bezos Expeditions, Samsung, LG Electronics, Hyundai Motor Group, Fidelity, and others, Tenstorrent has raised over $1B+ and operates from Santa Clara, Austin, Toronto, Belgrade, Tokyo, and Bangalore.
pr@tenstorrent.com
About Smallest.ai
Smallest.ai is a foundational AI research lab building the next generation of real-time voice AI infrastructure for enterprises. The company develops speech recognition, speech generation, and speech-to-speech systems designed to enable natural, scalable AI conversations across customer service, healthcare, financial services, and other high-volume communication environments. Headquartered in San Francisco, Smallest.ai serves enterprises globally through its Voice 4.0 platform and proprietary AI models. The company is backed by Seligman Ventures, Sierra Ventures, 3one4 Capital, Better Capital, Upsparks Capital, Schema Ventures, Tiny VC, DeVC, Mission Street Capital, and other angel investors.
AnnouncementsOpen SourceNewsroom MiniMax M3.1 Flash, Now Live on Tenstorrent
Oct 22, 2026
AnnouncementsNewsroomArchitecture Tenstorrent Releases Open Source Beta of the Open Chiplet Atlas® (OCA) Harness
Oct 22, 2026
AnnouncementsNewsroom PFN and Tenstorrent Selected for NEDO Project on “xPU Design and Development Incorporating Photonics Electronics Integration Technology”
Oct 21, 2026