llama : synchronize context before backend teardown - #24935

llama : synchronize context before backend teardown#24935

ggerganov merged 1 commit into

ggml-org:masterggml-org/llama.cpp:masterfrom

krystophny:fix/cuda-ctx-synchronize-on-destroykrystophny/llama.cpp:fix/cuda-ctx-synchronize-on-destroyCopy head branch name to clipboard