What to Know About Your Workstation

Your workstation is a Tenstorrent Quietbox 2: four AI accelerators inside, an operating system you may not have used before, and the software stack already configured and waiting. The machine is ready to go — what’s left is knowing what you’ve got.

This guide doesn’t assume you know Linux, or Python, or what a PCIe slot is. It assumes you’re curious, and that curiosity is enough.

What’s Inside

The Tenstorrent Quietbox 2 (QB2) is a workstation with two Blackhole p300c cards — four Blackhole chips in total — on PCIe. Each p300c is a dual-chip card, and each chip is independent — four separate devices from the software’s point of view, connected to a standard CPU running Ubuntu 24.04 LTS.

What	Detail
AI chips	2× Blackhole p300c cards (4 Blackhole chips)
Tensix cores per chip	120 (12×10 compute grid)
Connection	PCIe Gen4 (4 independent devices)
OS	Ubuntu 24.04 LTS
Pre-installed	TTNN, vLLM, tt-smi, drivers, Python venvs
Source tree	Not included — `~/tt-metal` has venvs, not source

The chips don’t share memory. When you open device 0, you’re talking to one Blackhole chip. To use all four together, you use ttnn.CreateDevices({0, 1, 2, 3}) — not four separate open_device() calls.

⬡ Each Blackhole chip is a 17×12 Network on Chip (NoC) grid — 204 positions in total. Of those, 140 are Tensix compute tiles (120 are enabled on QB2's chips; the rest are harvested); the remainder are DRAM controllers, Ethernet cores for chip-to-chip links, the PCIe interface, and the routing fabric between them. The grid is how work moves — not through a shared bus, but through a programmable mesh of message-passing nodes.

Four Blackhole chips connected via PCIe to the CPU, with software stack and Python environments

🔌 Before anything else: power switch on the back panel to the ON position, then press the front power button. The fans spin up. That's the QB2 waking up. That sound is correct and expected.

What Ships Pre-Installed

Tenstorrent ships the QB2 ready to serve models. You don’t install drivers. You don’t compile anything. The full stack is already there:

Kernel driver — loaded automatically at boot, makes the chips visible to software
tt-smi — hardware monitoring tool, lives at /usr/bin/tt-smi
TTNN Python environment — pre-built venv at ~/tt-metal/python_env/
vLLM — in the main tenstorrent venv at ~/.tenstorrent-venv/
TT-Forge/XLA — container wrapper at ~/.local/bin/tt-forge
tt-studio — the no-code web UI for serving models, pre-installed (launch with tt-studio)
A ready-to-run model — Qwen3-32B, weights pre-cached on disk, deployable from tt-studio with no download
Firmware — already flashed to all four chips

What’s intentionally absent: the ~/tt-metal source code. The environments are there; the source isn’t. You can build models, run inference, and work with the full API stack without it. Building from source is a later chapter — a much later chapter.

Physical Tour

The QB2 looks like a standard tower workstation. On the inside:

CPU and motherboard running Ubuntu 24.04 LTS
Two Blackhole p300c cards (four Blackhole chips total)
RAM sized for production inference workloads
Storage for model weights — but watch it carefully (more on that in Chapter 2)

The chips run warm under load. Fans will get louder when you run inference. This is correct. The cooling is designed for sustained operation at full chip temperature.

⬡ Tensix Grid — Blackhole (P100/P150/P300c / QB2)

One Blackhole chip. You have four, on two p300c cards.

Next: First Boot →