A pure C inference engine that runs very large mixture-of-experts models on ordinary hardware by streaming routed experts from disk.
42,8 mil4,8 milCApache-2.0Actualizado hace 3 horas
#vulkan
Repositorios de GitHub seleccionados relacionados con vulkan, con resúmenes de IA y métricas clave.
A pure C inference engine that runs very large mixture-of-experts models on ordinary hardware by streaming routed experts from disk.