tqcq/llama.cpp
0
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2025-06-29 04:35:05 +00:00
Code Issues Packages Projects Releases Wiki Activity

embedding : avoid common_batch

Browse Source 
ggml-ci
This commit is contained in:
Georgi Gerganov
2025-03-19 14:29:04 +02:00
parent 6f54ee660c
commit 8b80d68338
2 changed files with 33 additions and 36 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
examples/retrieval/retrieval.cpp
View File

@ -82,7 +82,7 @@ static void batch_add_seq(llama_batch_ext * batch, const std::vector<int32_t> &
}
static void batch_decode(llama_context * ctx, llama_batch_ext * batch, float * output, int n_seq, int n_embd, int embd_norm = 2) {
const struct llama_model * model = llama_get_model(ctx);
const llama_model * model = llama_get_model(ctx);
// clear previous kv_cache values (irrelevant for embeddings)
llama_kv_self_clear(ctx);