Development repository for the Triton language and compiler

You like pytorch? You like micrograd? You love tinygrad! ❤️

Vx: one language, every chip.

Website for graphing performance of rustc

Rust's standard library vendor-specific APIs and run-time feature detection

Personal blog

eLLM: Run Long-Horizon Inference Faster on CPUs Than on GPUs

Companion code for the global workspace interpretability paper

An eval-aware model organism that never says it is being evaluated

This is an experiment on induction heads across three different models i.e. GPT-2 small, Pythia-160m and Qwen2.5-0.5B. It involves knowing what it cost for these models when the induction heads are…

The nnsight package enables interpreting and manipulating the internals of deep learned models.

Examples and support libraries for the amdgpu Rust target

A modern runtime for JavaScript and TypeScript.

A runtime for writing reliable asynchronous applications with Rust. Provides I/O, networking, scheduling, timers, ...

A high-throughput and memory-efficient inference and serving engine for LLMs

Rust's standard library vendor-specific APIs and run-time feature detection

Moved to Codeberg

Wayland native area selector for compositors implementing zwlr_layer_shell.

A secure, scalable transaction service built with Rust and Axum for managing business accounts, processing financial transactions, and delivering webhooks.

a model of MIR and the Rust type/trait system

An extremely fast Python package and project manager, written in Rust.

Format Rust code

Cargo file formatter