CS348Project/llama.cpp
5
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2025-11-05 09:36:52 +00:00
Code Issues Packages Projects Releases Wiki Activity

llama : add new llama_decode() API that works with llama_batch

Browse Source
This commit is contained in:
Georgi Gerganov
2023-09-18 14:23:52 +03:00
parent 58bb5110ca
commit 9f42e75489
13 changed files with 146 additions and 75 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
common/common.cpp
View File

@@ -780,7 +780,7 @@ std::tuple<struct llama_model *, struct llama_context *> llama_init_from_gpt_par
LOG("warming up the model with an empty run\n");
std::vector<llama_token> tmp = { llama_token_bos(lctx), llama_token_eos(lctx), };
llama_eval(lctx, tmp.data(), std::min(tmp.size(), (size_t) params.n_batch), 0, params.n_threads);
llama_decode(lctx, llama_batch_get_one(tmp.data(), std::min(tmp.size(), (size_t) params.n_batch), 0, 0), params.n_threads);
llama_reset_timings(lctx);
}