# tracel-ai/cubecl

Multi-platform high-performance compute language extension for Rust.

Repository: https://github.com/tracel-ai/cubecl
Canonical: https://ross.abutalabs.com/products/cubecl
Homepage: https://burn.dev
Language: Rust
License: Apache-2.0
License Family: permissive
Topics: cuda, gpgpu, gpu, jit, linalg, rust, webgpu
Last push: 2026-08-26T21:24:21+00:00

## Health v2 (maintenance only)
Score: 81/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 99, release rhythm 71, longevity 57
- inputs: {"age_days": 810, "days_push": 7, "days_rel": 114, "gap_med": 85, "n_releases_24m": 8}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 2334, forks 230 (observed 2026-08-28T04:06:37.819601+00:00)

## What it is
CubeCL is a Rust language extension, JIT compiler, and runtime set for writing high-performance GPU compute kernels in plain Rust. A single #[cube] function compiles on demand to CUDA, HIP, Metal, SPIR-V, WGSL, or CPU SIMD with compile-time specialization and autotuning.

## Use cases
- write gpu compute kernels in rust
- compile one kernel to cuda and webgpu
- use tensor cores from rust
- build a multi-backend linalg library
- avoid writing wgsl or cuda c by hand
- autotune kernel launch configurations

## When to choose
- you need peak-performance kernels across nvidia, amd, apple, and webgpu backends
- you want type-checked, testable kernels in rust instead of shader strings
- you're building scientific computing or deep learning infrastructure in rust

## When to avoid
- you just need simple graphics rendering rather than compute kernels
- you want a high-level deep learning framework rather than a kernel layer
- your project doesn't use rust

## Facets
- artifact type: library
- maturity: active
- function: compiler, gpu-computing, machine-learning, math
- domain: gpu-computing, machine-learning, programming-languages, compilers, performance
- platform: rust, cross-platform, wasm
- tags: jit-compiler, cuda, webgpu, gpgpu, compute-kernels, tensor-cores, autotuning, rust-language-extension, gpu

## Member repositories
- tracel-ai/cubecl (main) score 81

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:06:37.819601+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T02:37:54.265311+00:00, confidence not recorded.
  - readme: https://github.com/tracel-ai/cubecl (fetched 2026-08-28T04:06:37.819601+00:00, sha 28a6ae4eaa63)
  - homepage: https://burn.dev (fetched 2026-08-29T10:18:14.698741+00:00, sha 2991f31e852e)
  - registry_crates: https://crates.io/api/v1/crates/cubecl (fetched 2026-08-29T10:18:14.707611+00:00, sha eee34a5f7747)
- Data as of 2026-08-30T08:39:29.467469+00:00.