ver4a/llama.cpp
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity

llama : rework embeddings logic (#14208)

Browse source 
* llama : rework embeddings logic

ggml-ci

* cont : fix rerank

ggml-ci

* cont : engrish [no ci]

* cont : fix rerank

ggml-ci

* server : support both embeddings and completions with single model

ggml-ci

* cont : avoid embeddings_org

ggml-ci
This commit is contained in:
Georgi Gerganov 2025-06-16 14:14:00 +03:00 • committed by GitHub
parent 3ba0d843c6
commit d3e64b9f49
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
16 changed files with 159 additions and 114 deletions
Download patch file Download diff file Expand all files Collapse all files

12
include/llama.h
View file

@ -254,7 +254,10 @@ extern "C" {
// - seq_id : the sequence to which the respective token belongs
// (if set to NULL, the sequence ID will be assumed to be 0)
// - logits : if zero, the logits (and/or the embeddings) for the respective token will not be output
// (if set to NULL, only the logits for last token will be returned)
// (if set to NULL:
// - if embeddings: all tokens are output
// - if not: only the last token is output
// )
//
typedef struct llama_batch {
int32_t n_tokens;
@ -262,8 +265,8 @@ extern "C" {
llama_token * token;
float * embd;
llama_pos * pos;
int32_t * n_seq_id; // TODO: remove, should belong to only 1 sequence
llama_seq_id ** seq_id; // TODO: become llama_seq_id * seq_id;
int32_t * n_seq_id;
llama_seq_id ** seq_id;
int8_t * logits; // TODO: rename this to "output"
} llama_batch;
@ -961,8 +964,7 @@ extern "C" {
// Get the number of threads used for prompt and batch processing (multiple token).
LLAMA_API int32_t llama_n_threads_batch(struct llama_context * ctx);
// Set whether the model is in embeddings mode or not
// If true, embeddings will be returned but logits will not
// Set whether the context outputs embeddings or not
LLAMA_API void llama_set_embeddings(struct llama_context * ctx, bool embeddings);
// Set whether to use causal attention or not