Dense Compute for Real Workloads
Run hundreds of cutting-edge open-weight models – all locally – including LLMs and generative models up to 120B parameters. Develop generative image and video, speech-to-text, text-to-speech, and computer vision models at your desk.
Plug In & Start Developing
All-in-one workstation for your home, office, or lab. Plug in with a standard power outlet, no rack needed. Arrives with pre-built enterprise quality system software stack and dependencies installed.
Open Source
Edit, select, fork, and own your AI with our open source full stack software. Get started quickly compiling models with TT-Forge. For hardware optimization and kernel development, check out TT-Metalium, our low level AI SDK, and TT-LLK, our low level kernel software.
Right-Next-to-You Local AI
TT-QuietBox™ 2 is a whisper-quiet liquid-cooled AI workstation delivering multi-chip ASIC compute with an open-source software stack that can plug-in at your desk. Run up to 120B parameter models locally, accelerate smaller models across multiple chips, or run inference workloads without cloud dependency. Plug-in, boot up, and start developing immediately.
TT-QuietBox™ 2 is coming Q2 2026!
Be the first to know. Join the waitlist for updates.
*Required
