public_git/whisper.cpp
1
0
Fork 0
mirror of https://github.com/ggml-org/whisper.cpp.git synced 2026-10-18 21:55:18 +02:00
Code Issues Packages Projects Releases Wiki Activity

talk-llama : sync llama.cpp

Browse Source
This commit is contained in:
Georgi Gerganov
2025-09-29 15:18:41 +03:00
parent 404a93114c
commit fcf0181ee2
21 changed files with 606 additions and 43 deletions
Download Patch File Download Diff File Expand all files Collapse all files
examples/talk-llama/llama-kv-cache.cpp
+8
View File
@@ -473,6 +473,14 @@ llama_pos llama_kv_cache::seq_pos_max(llama_seq_id seq_id) const {
return cells.seq_pos_max(seq_id);
}
std::map<ggml_backend_buffer_type_t, size_t> llama_kv_cache::memory_breakdown() const {
std::map<ggml_backend_buffer_type_t, size_t> ret;
for (const ggml_backend_buffer_ptr & buf_ptr : bufs) {
ret[ggml_backend_buffer_get_type(buf_ptr.get())] += ggml_backend_buffer_get_size(buf_ptr.get());
}
return ret;
}
llama_memory_context_ptr llama_kv_cache::init_batch(
llama_batch_allocr & balloc,
uint32_t n_ubatch,