Files
candle/candle-kernels
Laurent Mazare e3db30021f Support for "unbatched" rope. (#2926)
* Support for (un)-batched rope.

* Use 3d rope in the rope/ropei/rope_thd functions.

* Get the CPU versions to work.

* Fix the cuda version.

* Adapt the metal side.

* Fix the metal tests.
2025-04-27 15:12:02 +02:00
..
2025-04-27 15:12:02 +02:00

candle-kernels

This crate contains CUDA kernels used from candle. Some of these implementations come from the dfdx crate.