mlx/python
Ethan a749a91c75
Support disable metal buffer cache to prevent performance degradation caused by large memory caching (#390)
* support disable metal buffer cache, due to large unused memory buffered when llm generated long context tokens

* Run format and add "cache_enabled" feature tests
2024-01-18 08:33:34 -08:00
..
mlx Add Gaussian NLL loss function (#477) 2024-01-18 06:44:44 -08:00
src Support disable metal buffer cache to prevent performance degradation caused by large memory caching (#390) 2024-01-18 08:33:34 -08:00
tests Added formatter structure and a boolean value formatter (#354) 2024-01-18 07:49:41 -08:00