tqcq/llama.cpp
0
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2025-06-28 20:25:20 +00:00
Code Issues Packages Projects Releases Wiki Activity

llama_batch_ext_ptr::from_text/embd

Browse Source
This commit is contained in:
Xuan Son Nguyen
2025-03-14 17:12:03 +01:00
parent 8e7714fa77
commit eaffba0f2e
12 changed files with 34 additions and 14 deletions
Download Patch File Download Diff File Expand all files Collapse all files

22
include/llama-cpp.h
View File

@ -32,4 +32,24 @@ typedef std::unique_ptr<llama_model, llama_model_deleter> llama_model_ptr;
typedef std::unique_ptr<llama_context, llama_context_deleter> llama_context_ptr;
typedef std::unique_ptr<llama_sampler, llama_sampler_deleter> llama_sampler_ptr;
typedef std::unique_ptr<llama_adapter_lora, llama_adapter_lora_deleter> llama_adapter_lora_ptr;
typedef std::unique_ptr<llama_batch_ext, llama_batch_ext_deleter> llama_batch_ext_ptr;
struct llama_batch_ext_ptr : std::unique_ptr<llama_batch_ext, llama_batch_ext_deleter> {
llama_batch_ext_ptr(llama_batch_ext * batch) : std::unique_ptr<llama_batch_ext, llama_batch_ext_deleter>(batch) {}
// convience function to create a batch from text tokens, without worrying about manually freeing it
static llama_batch_ext_ptr from_text(llama_token * tokens,
int32_t n_tokens,
int32_t pos0,
int32_t seq_id,
bool output_last) {
return llama_batch_ext_ptr(llama_batch_ext_init_from_text(tokens, n_tokens, pos0, seq_id, output_last));
}
static llama_batch_ext_ptr from_embd(float * embd,
size_t n_tokens,
size_t n_embd,
int32_t pos0,
int32_t seq_id) {
return llama_batch_ext_ptr(llama_batch_ext_init_from_embd(embd, n_tokens, n_embd, pos0, seq_id));
}
};