6 releases
Uses new Rust 2024
| new 0.2.0-pre.2 | Mar 2, 2026 |
|---|---|
| 0.2.0-pre.1 | Feb 9, 2026 |
| 0.1.1 | Jan 23, 2026 |
| 0.1.0-pre.1 | Dec 18, 2025 |
| 0.0.1 | Dec 5, 2025 |
#2249 in Math
21,233 downloads per month
Used in 44 crates
(4 directly)
35KB
502 lines
CubeK: high-performance multi-platform kernels in CubeCL
Algorithms
| Algorithms | Variants |
|---|---|
| Random | bernoulli normal uniform |
| Quantization | symmetric per-block per-tensor q2 q4 q8 fp4 |
| Reduction | mean sum prod max min arg[max|min] per-cube per-plane |
| Matmul | mma unit tma multi-stage specialization ordered multi-rows |
| Convolution | mma unit tma multi-stage im2col |
| Attention | mma unit multi-rows |
Contributing
If you want to contribute new kernels, please read the GUIDE.md.
Dependencies
~62–105MB
~2M SLoC