tenstorrenttenstorrentOfficial·5 Agent Skills Included

tt-mlir

Compile and optimize ML models for Tenstorrent hardware

Compiles machine learning models into optimized binaries for Tenstorrent accelerators using MLIR dialects. Guides adding new compiler ops, decomposing fused operations, and lowering code across TTIR, TTNN, and D2M layers. Automates cross-repo PR validation, simulator CI uplifts, and op-by-op compile testing to catch regressions early.
npx skills add tenstorrent/tt-mlir --all -g -y
Available:

Gives your AI agent the exact build, lint, and test commands plus coding rules needed to safely modify and verify changes in this compiler project.

All Skills in This Repository (5)

Pure Emerald Level Indicators

Frequently Asked Questions

FAQPage Schema
How to install tt-mlir?

Run `npx skills add tenstorrent/tt-mlir --all -g -y` in your terminal to install all skills in this suite globally.

What does tt-mlir do?

It is an open-source compiler that translates machine learning models into optimized binaries for Tenstorrent AI hardware like Wormhole and Blackhole.

How to add a new op to tt-mlir?

The add-op skill walks your agent through every layer, from dialect definitions and conversions to flatbuffer serialization, runtime code, and tests.

Can I validate a tt-mlir PR against tt-xla?

Yes. The validation skill cherry-picks your PR onto the pinned commit, triggers tt-xla CI, and reports pass or failure results back to the PR.

Do I need compiler experience to use these skills?

No. The skills give your agent step-by-step instructions, file locations, and code templates so it handles the compiler details for you.

Related Repositories in Software Engineering

View All in Software Engineering