cturan/llama.cpp

miroir de https://github.com/cturan/llama.cpp

Auteur	SHA1 Message	Date
Georgi Gerganov	cd5e3b5754 server : support unified cache across slots (#16736)	il y a 2 mois
Acly	d8eaa26e4d tests : fix test-thread-safety when compiling with multiple backends (#16699)	il y a 2 mois
Georgi Gerganov	00131d6eaf tests : update for LLAMA_SET_ROWS=1 (#14961)	il y a 5 mois
Diego Devesa	6adc3c3ebc llama : add thread safety test (#14035)	il y a 7 mois