From f482bb2e4920e544651fb832f2e0bcb4d2ff69ab Mon Sep 17 00:00:00 2001 From: Pierrick Hymbert Date: Sat, 23 Mar 2024 18:07:00 +0100 Subject: common: llama_load_model_from_url split support (#6192) * llama: llama_split_prefix fix strncpy does not include string termination common: llama_load_model_from_url: - fix header name case sensitive - support downloading additional split in parallel - hide password in url * common: EOL EOF * common: remove redundant LLAMA_CURL_MAX_PATH_LENGTH definition * common: change max url max length * common: minor comment * server: support HF URL options * llama: llama_model_loader fix log * common: use a constant for max url length * common: clean up curl if file cannot be loaded in gguf * server: tests: add split tests, and HF options params * common: move llama_download_hide_password_in_url inside llama_download_file as a lambda * server: tests: enable back Release test on PR * spacing Co-authored-by: Georgi Gerganov * spacing Co-authored-by: Georgi Gerganov * spacing Co-authored-by: Georgi Gerganov --------- Co-authored-by: Georgi Gerganov --- common/common.h | 7 +++++++ 1 file changed, 7 insertions(+) (limited to 'common/common.h') diff --git a/common/common.h b/common/common.h index afa4cf6d..a223ecea 100644 --- a/common/common.h +++ b/common/common.h @@ -306,3 +306,10 @@ struct llama_control_vector_load_info { // Load control vectors, scale each by strength, and add them together. // On error, returns {-1, empty} llama_control_vector_data llama_control_vector_load(const std::vector & load_infos); + +// +// Split utils +// +static const char * const LLM_KV_SPLIT_NO = "split.no"; +static const char * const LLM_KV_SPLIT_COUNT = "split.count"; +static const char * const LLM_KV_SPLIT_TENSORS_COUNT = "split.tensors.count"; -- cgit v1.2.3