tqcq/llama.cpp
0
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2025-06-29 12:35:16 +00:00
Code Issues Packages Projects Releases Wiki Activity

fix llama_batch_ext_init_from_text

Browse Source
This commit is contained in:
Xuan Son Nguyen
2025-03-13 23:09:27 +01:00
parent 65f0184517
commit c3dd79007b
18 changed files with 40 additions and 27 deletions
Download Patch File Download Diff File Expand all files Collapse all files

6
examples/speculative/speculative.cpp
View File

@ -166,9 +166,9 @@ int main(int argc, char ** argv) {
const auto t_enc_start = ggml_time_us();
// eval the prompt with both models
llama_batch_ext_ptr batch0(llama_batch_ext_init_from_text( inp.data(), n_input - 1, 0, 0));
llama_batch_ext_ptr batch1(llama_batch_ext_init_from_text(&inp.back(), 1, 0, 0));
llama_batch_ext_ptr batch2(llama_batch_ext_init_from_text( inp.data(), n_input , 0, 0));
llama_batch_ext_ptr batch0(llama_batch_ext_init_from_text( inp.data(), n_input - 1, 0, 0, true));
llama_batch_ext_ptr batch1(llama_batch_ext_init_from_text(&inp.back(), 1, 0, 0, true));
llama_batch_ext_ptr batch2(llama_batch_ext_init_from_text( inp.data(), n_input , 0, 0, true));
llama_decode_ext(ctx_tgt, batch0);
llama_decode_ext(ctx_tgt, batch1);
llama_decode_ext(ctx_dft, batch2);