Developers

Learn how to get your models up and running fast on Tenstorrent hardware. With two open source SDKs, you can get as close to the metal as possible, or let our AI compiler do the work.
TT-Metalium™ GitHub
TT-Forge GitHub
Documentation Site
Developers

Model Support Table

Qwen 3 32B
QuietBox (Wormhole)
LLM
TP=8
QwQ 32B
QuietBox (Wormhole)
LLM
TP=8
DeepSeek R1 Distill Llama 3.3 70B
QuietBox (Wormhole)
LLM
TP=8
Llama 3.1 70B
Galaxy
LLM
TP=32
Llama 3.1 70B
QuietBox (Wormhole)
LLM
TP=8
Llama 3.1 70B
QuietBox (Blackhole)
LLM
TP=4
Llama 3.2 11B Vision
n300
LLM
TP=2
Qwen 2.5 7B
n300
LLM
TP=2
Qwen 2.5 72B
QuietBox (Wormhole)
LLM
TP=8
Falcon 7B
n150
LLM
Falcon 7B
QuietBox (Wormhole)
LLM
DP=8
Falcon 7B
Galaxy
LLM
DP=32
Falcon 40B
QuietBox (Wormhole)
LLM
TP=8
Llama 3.1 8B
p100
LLM
Llama 3.1 8B
p150
LLM
Llama 3.1 8B
2 x p150
LLM
DP=2
Llama 3.1 8B
n150
LLM
Llama 3.2 1B
n150
LLM
Llama 3.2 3B
n150
LLM
Mamba 2.8B
n150
LLM
Mistral 7B
n150
LLM
Mixtral 8x7B
QuietBox (Wormhole)
LLM
TP=8
BERT-Large
n150
NLP
Sentence-Bert (backbone: bert-base)
n150
NLP

Getting started on Tenstorrent

Wormhole™ n150s & n300s

Doc Site

TT-QuietBox Workstation

Doc Site

TT-LoudBox Workstation

Doc Site
Looking for other documentation?
See All Docs
Active Bounties
Solve bugs and add features to win cash prizes, and get our open source software to stable releases even faster.
Terms and Conditions
unknown
tt-metal builds on rocky linux 9
unknown
Use exp_init/recip_init instead of hardcoded constants, rsqrt_tile instead of sqrt_tile+recip_tile.
hard
[$2500 BOUNTY] Bring Up TT-Metal on Asahi
$2500
See All Bounties
Upcoming Events
See All
Jul 23
Building AI agents with Tenstorrent

Tenstorrent hardware is designed to optimize the operations that power AI. Learn how you can run models on Tenstorrent hardware to build multi-agent systems and workflows.

Learn more
Aug 1
AI Meetup Malaysia

​Join Tenstorrent and AWS and learn about AI with an emphasis on computer architecture, MLOps, and models.

Learn more
Aug 9
COSCUP 2025

Stop by our booth at the Conference for Open Source Coders, Users & Promoters (COSCUP), the largest open source conference in Asia.

Learn more

Educational Content

Tutorials

See all
Intro to TT-Forge
Tutorials
Intro to TT-Forge
An overview of TT-Forge, Tenstorrent's MLIR-based compiler.
TT-Metalium Programming Overview
Tutorials
TT-Metalium Programming Overview
A guide to the Metalium programming model, Tenstorrent’s primary programming model.
How to Use the TTNN Visualizer
Tutorials
How to Use the TTNN Visualizer
Learn how to install and run ttnn-visualizer. ttnn-visualizer helps you gain a complete understanding of a model.

Written Tutorials

See all
Bring up LLMs with TTNN
Get guidance on how to bring up high-performance multi-chip models on Tenstorrent hardware using the TT-Metalium stack.
Read article
Get started with TTNN-Visualizer
A quickstart guide to setting up ttnn-visualizer.
Read article
Op Writer's Guide to Dispatch Overhead
This tutorial covers different methods to optimize dispatch overhead resource allocation, kernel initialization, and runtime arguments.
Read article

Frequently Asked Questions

View all
Installation
Installation and running matmul_single_core example
General
Is computation run on a Tenstorrent Tensix core accelerator card guaranteed to be deterministic?
General
How to Reset an Accelerator Board

Join the Community

Get access to support on anything from setting up new hardware, running models or optimizing your setup, plus the latest on Tenstorrent hardware and software.
Join the Discord
Community

Interested in contributing?

Tenstorrent's AI software stack is open source. Getting started is as easy as filing an issue.
Learn More