Convert PyTorch tensors and execute tensor operations on Tenstorrent MeshDevice topologies.
npx skills add https://github.com/tenstorrent/tt-forge --skill ttnn
Please help me install this Agent Skill. Skill: ttnn Source: https://github.com/tenstorrent/tt-forge/tree/main/skills/ttnn Command: npx skills add https://github.com/tenstorrent/tt-forge --skill ttnn
TTNN provides high-performance tensor primitives and device management on Tenstorrent hardware, enabling efficient ML workloads and model conversions.
Initialize TTNN on a device, convert a PyTorch tensor to TTNN, and execute a simple tensor operation on a MeshDevice.
High-intent search queries and answers about installing and using this skill.
To convert a PyTorch model to TTNN, use the provided model conversion utilities to transform PyTorch tensors into TTNN compatible formats for accelerated inference on Tenstorrent hardware.
A MeshDevice topology enables multi-device execution in TTNN by managing tensor operations and memory configurations across distributed Tenstorrent hardware for scalable ML workloads.
You need knowledge of the TT-Metal stack and MLIR pipelines to use TTNN APIs effectively, as TTNN integrates with TT-Metal for low-level tensor operations and TT kernels.
TTNN provides high-level APIs for tensor creation, manipulation, and movement across MeshDevice topologies, requiring specific memory configurations to manage execution on Tenstorrent hardware.
The best way to scale tensor operations is using TTNN's multi-device execution across MeshDevice topologies, which handles tensor movement and memory configurations for distributed ML workloads.
tenstorrentEstablish remote connections to Tenstorrent hardware and run kernels, copy files, and read logs.
Clay-HHKProcess large-scale ML datasets across clusters for distributed preprocessing and batch inference.
tenstorrentWrite TT hardware kernels using a Python DSL with data movement primitives.
moritztngPort biomolecular models to Tenstorrent hardware within the tt-bio stack.
