CS348Project/llama.cpp
5
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2025-10-31 08:51:55 +00:00
Code Issues Packages Projects Releases Wiki Activity

fix common_batch missing seq_id

Browse Source
This commit is contained in:
Xuan Son Nguyen
2025-03-13 22:38:04 +01:00
parent 47086fa82d
commit 9fb2d81eab
1 changed files with 3 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files

5
common/common.h
View File

@@ -586,6 +586,7 @@ struct common_batch {
llama_batch_ext_ptr batch;
struct batch_token {
llama_token token;
llama_seq_id seq_id; // only support single seq for now
bool logits;
};
std::vector<batch_token> tokens;
@@ -601,14 +602,14 @@ struct common_batch {
}
void add_text(llama_token token, llama_pos pos, llama_seq_id seq_id, bool logits) {
llama_batch_ext_add_text(batch.get(), token, pos, &seq_id, 1, logits);
tokens.push_back({token, logits});
tokens.push_back({token, seq_id, logits});
if (logits) {
n_outputs++;
}
}
void add_text(llama_token token, llama_pos pos, std::vector<llama_seq_id> seq_ids, bool logits) {
llama_batch_ext_add_text(batch.get(), token, pos, seq_ids.data(), seq_ids.size(), logits);
tokens.push_back({token, logits});
tokens.push_back({token, seq_ids[0], logits});
if (logits) {
n_outputs++;
}