Liquid-Cooled Desktop System Runs Models up to 120B Parameters Locally With a Fully Open-Source Stack, Starting at $9,999

SANTA CLARA, CA / ACCESS Newswire / March 11, 2026 / Tenstorrent, the AI computing company led by CEO Jim Keller, today announced TT-QuietBox™ 2 (Blackhole™). This whisper-quiet, liquid-cooled AI workstation runs models up to 120 billion parameters directly at your desk, ships with an entirely open-source software stack from compiler to kernel, and starts at $9,999. It marks the industry's first desktop AI workstation built on RISC-V architecture to deliver teraflop-class inference.

The Inference Imperative

The timing 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 - and it is still accelerating. 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.

QuietBox 2 is built around a different proposition: Developers doing the actual work of AI should be able to see, control, and own every layer of their compute - from silicon architecture to the compiler. It is ideal for developers and small to medium business deployments requiring on prem deployment without the racks.

"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," said Jim Keller, CEO of Tenstorrent.

Real Workloads Out of the Box

QuietBox 2 ships ready for quick deployment. It excels across diverse AI domains:

LLMs & Coding: GPT-OSS 120B runs entirely on-device - a full 120-billion-parameter model operating privately at your desk. Llama 3.1 70B runs at 476.5 tokens per second. Qwen3-32B deploys as a private coding agent, reasoning through entire codebases without cloud token limits.
Creative & Multimodal: Flux handles image generation and Wan 2.2 handles video synthesis entirely locally, ensuring creative IP remains off third-party servers.
Scientific Research: Boltz-2, a biomolecular ML model, predicts the structure of a 686-amino-acid protein in just 49 seconds on a single Blackhole processor - a task that takes a modern CPU 45 minutes. This matches the performance of flagship workstation GPUs at a fraction of the cost. QuietBox 2 can predict four protein structures in parallel, yielding 4x higher throughput.

For models not on the pre-installed list, TT-Forge - Tenstorrent's open-source AI compiler - can run models from PyTorch, ONNX, TensorFlow, JAX, and PaddlePaddle directly to the hardware. If it runs on a standard framework, it can run on QuietBox 2.

Silicon Innovation Without Memory + Networking Bottlenecks

Four Blackhole ASICs work as a unified mesh inside a single desk-friendly enclosure. The system features 480 Tensix cores delivering 2,654 TFLOPS at BlockFP8 precision, backed by 128 GB of GDDR6 high-speed memory and 256 GB of DDR5 system memory.

This architecture integrates compute and high-density SRAM on a single die. This dataflow approach moves tensors efficiently through on-chip memory, completely sidestepping the DRAM bottlenecks that limit sustained throughput on conventional hardware. By utilizing GDDR6 and on-chip SRAM, QuietBox 2 entirely avoids the High-Bandwidth Memory (HBM) supply shortages currently driving price hikes across the rest of the AI hardware market.

The system runs on Ubuntu 24.04, plugs into a standard 120V wall outlet, and requires no rack, specialized electrical work, or server room.

Open Source at Every Layer

Every layer of QuietBox 2's software is open source. This is not just an open API on a black box; it is full stack visibility.

TT-Forge gives developers total visibility into graph lowering, transformation, optimization, and execution.
TT-Metalium, the low-level AI SDK, provides kernel-level control with deterministic execution.
TT-LLK handles low-level kernel software.

Developers can see exactly what happens at every stage of their pipeline, debug at the hardware level, fork any component, and modify the stack to fit their exact workload. For sovereign AI deployments, regulated industries, and research institutions that must guarantee how their infrastructure handles data, this transparency is not just a feature - it is the core architecture.

A Fun Developer Experience

QuietBox 2 represents a ground-up redesign focused on developer velocity and environmental efficiency. The system ships fully pre-configured with Ubuntu 24.04, the complete open-source software stack, and TT-Studio, enabling quick deployment right out of the box.

Engineering advancements have reduced idle power consumption and heat output by approximately 50% compared to previous generations. Coupled with significantly expanded documentation and developer tooling, the new liquid-cooled chassis is engineered specifically for quiet, sustained, heavy-workload operation directly on a desk.

For small and medium businesses, the same plug-and-play simplicity translates to on-premises AI deployment without dedicated server rooms, specialized electrical work, or IT overhead - inference capacity that previously required a rack, delivered in a desktop form factor.

Availability: TT-QuietBox™ 2 ships globally in Q2 2026, starting at $9,999. To join the waitlist, visit www.tenstorrent.com/waitlist/tt-quietbox.

The system will be demonstrated live at the Game Developers Conference (GDC) 2026, March 11-13, at Tenstorrent's booth #1354. To schedule a press meeting at GDC, contact Salient PR at the address below.

About Tenstorrent

Tenstorrent is an AI compute company. Led by CEO Jim Keller - architect of AMD Zen, Apple A4/A5, and Tesla's Full Self-Driving chip - the company builds RISC-V-based AI processors and systems for developers, enterprises, and sovereign infrastructure worldwide. Backed by Bezos Expeditions, Samsung, LG Electronics, Hyundai Motor Group, Fidelity, and others, Tenstorrent has raised over $1 billion and operates from Santa Clara, Austin, Toronto, Belgrade, Tokyo, Bangalore, Singapore, and Seoul. Learn more at tenstorrent.com.

Media Contact:
Justin Mauldin
Salient PR
achievemore@salientpr.com
737.234.0936

SOURCE: Tenstorrent

View the original press release on ACCESS Newswire

Logo

Tags

More Like This

Positron AI Raises $230 Million Series B at Over $1 Billion Valuation to Scale Energy-Efficient AI Inference

ASUS Announces Availability of ASUS Ascent GX10 Personal AI Supercomputer

Oxmiq Labs Inc.™: Re-Architecting the GPU Stack: From Atoms to Agents™

Quantinuum Selected as a Founding Collaborator for NVIDIA Accelerated Quantum Research Center

PsiQuantum Raises $1 Billion to Build Million-Qubit Scale, Fault-Tolerant Quantum Computers

GIGABYTE Announces its Personal AI Supercomputer AI TOP ATOM Will be Available Globally on October 15

Quantinuum Launches Industry-First, Trapped-Ion 56-Qubit Quantum Computer, Breaking Key Benchmark Record

Fujitsu develops generative AI reconstruction technology for optimized and energy-efficient AI models based on Takane LLM