vbatts/llama.cpp
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

llama : update llama_kv_self API

Browse source 
ggml-ci
This commit is contained in:
Georgi Gerganov 2025-01-14 16:47:34 +02:00
parent fd05ab87aa
commit 17b363afd3
No known key found for this signature in database
GPG key ID: 449E073F9DC10735
30 changed files with 387 additions and 205 deletions
Download patch file Download diff file Expand all files Collapse all files

3
examples/lookup/lookup.cpp
View file

@ -35,7 +35,6 @@ int main(int argc, char ** argv){
llama_model * model = llama_init.model.get();
llama_context * ctx = llama_init.context.get();
llama_kv_cache * kv = llama_get_kv_cache(ctx);
const llama_vocab * vocab = llama_model_get_vocab(model);
@ -193,7 +192,7 @@ int main(int argc, char ** argv){
// KV cache management
// clean the cache of draft tokens that weren't accepted
llama_kv_cache_seq_rm(kv, 0, n_past, -1);
llama_kv_self_seq_rm(ctx, 0, n_past, -1);
common_batch_clear(batch_tgt);
common_batch_add(batch_tgt, draft[0], n_past, { 0 }, true);