- Today
- Holidays
- Birthdays
- Reminders
- Cities
- Atlanta
- Austin
- Baltimore
- Berwyn
- Beverly Hills
- Birmingham
- Boston
- Brooklyn
- Buffalo
- Charlotte
- Chicago
- Cincinnati
- Cleveland
- Columbus
- Dallas
- Denver
- Detroit
- Fort Worth
- Houston
- Indianapolis
- Knoxville
- Las Vegas
- Los Angeles
- Louisville
- Madison
- Memphis
- Miami
- Milwaukee
- Minneapolis
- Nashville
- New Orleans
- New York
- Omaha
- Orlando
- Philadelphia
- Phoenix
- Pittsburgh
- Portland
- Raleigh
- Richmond
- Rutherford
- Sacramento
- Salt Lake City
- San Antonio
- San Diego
- San Francisco
- San Jose
- Seattle
- Tampa
- Tucson
- Washington
Tenstorrent Launches TT-QuietBox 2: RISC-V AI Workstation
Liquid-Cooled Desktop System Runs Models up to 120B Parameters Locally With a Fully Open-Source Stack, Starting at $9,999
Mar. 12, 2026 at 3:16am
Got story updates? Submit your updates here. ›
Tenstorrent, the AI computing company led by CEO Jim Keller, has announced the TT-QuietBox™ 2 (Blackhole™), a whisper-quiet, liquid-cooled AI workstation that runs models up to 120 billion parameters directly at your desk. The system ships with an entirely open-source software stack from compiler to kernel and starts at $9,999, marking the industry's first desktop AI workstation built on RISC-V architecture to deliver teraflop-class inference.
Why it matters
Inference has quietly overtaken training as the dominant AI workload, now accounting for more than 55% of cloud AI infrastructure spending at $37.5 billion. Yet, developers running these workloads face a stark choice: pay per-token cloud fees that compound as usage scales, or buy hardware locked to proprietary stacks they cannot inspect, modify, or truly own. The TT-QuietBox 2 is built around a different proposition, allowing developers to see, control, and own every layer of their compute from silicon architecture to the compiler.
The details
The TT-QuietBox 2 is built around four Blackhole ASICs that work as a unified mesh, delivering 2,654 TFLOPS at BlockFP8 precision. It features 128 GB of GDDR6 high-speed memory and 256 GB of DDR5 system memory, allowing it to run diverse AI workloads like large language models, creative and multimodal tasks, and scientific research entirely on-device without cloud token limits. The system runs on Ubuntu 24.04, plugs into a standard 120V wall outlet, and requires no rack, specialized electrical work, or server room.
- TT-QuietBox™ 2 ships globally in Q2 2026.
The players
Tenstorrent
An AI computing company led by CEO Jim Keller.
Jim Keller
The CEO of Tenstorrent.
What they’re saying
“Tenstorrent is working hard on open source AI software and we wanted to build a teraflop development system that was easy to use in a lab or office, fast and quiet. It's open top to bottom including the mechanical engineering. Build your own software or hardware. You can own your AI future.”
— Jim Keller, CEO of Tenstorrent (Mirage News)
What’s next
The TT-QuietBox™ 2 will be demonstrated live at the Game Developers Conference (GDC) 2026, March 11-13, at Tenstorrent's booth #1354.
The takeaway
The TT-QuietBox 2 represents a significant step forward in making high-performance, open-source AI computing accessible to developers and small to medium businesses, allowing them to run diverse AI workloads locally without the constraints and costs of cloud-based solutions.
