CS348Project/llama.cpp
5
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2025-10-30 08:42:00 +00:00
Code Issues Packages Projects Releases Wiki Activity

parallel : increase the variability of the prompt lengths (#13927)

Browse Source 
ggml-ci
This commit is contained in:
Georgi Gerganov
2025-05-30 19:38:07 +03:00
committed by GitHub
parent df0c0c7d02
commit dd665cc9d4
2 changed files with 6 additions and 3 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
examples/parallel/README.md
View File

@@ -4,7 +4,7 @@ Simplified simulation of serving incoming requests in parallel
## Example ## Example
Generate 128 client requests (`-ns 128`), simulating 8 concurrent clients (`-np 8`). The system prompt is shared (`-pps`), meaning that it is computed once at the start. The client requests consist of 10 junk questions (`-j 10`) followed by the actual question. Generate 128 client requests (`-ns 128`), simulating 8 concurrent clients (`-np 8`). The system prompt is shared (`-pps`), meaning that it is computed once at the start. The client requests consist of up to 10 junk questions (`--junk 10`) followed by the actual question.
```bash ```bash
llama-parallel -m model.gguf -np 8 -ns 128 --top-k 1 -pps --junk 10 -c 16384 llama-parallel -m model.gguf -np 8 -ns 128 --top-k 1 -pps --junk 10 -c 16384

7
examples/parallel/parallel.cpp
View File

@@ -315,7 +315,10 @@ int main(int argc, char ** argv) {
} else { } else {
client.prompt += k_system; client.prompt += k_system;
} }
for (int i = 0; i < n_junk; ++i) {
const int n_junk_cur = rand() % n_junk;
for (int i = 0; i < n_junk_cur; ++i) {
const int r = rand() % k_questions.size(); const int r = rand() % k_questions.size();
client.prompt += "User:\n" + k_questions[r] + "\nAssistant:\n " + k_answers[r] + "\n"; client.prompt += "User:\n" + k_questions[r] + "\nAssistant:\n " + k_answers[r] + "\n";
} }
@@ -340,7 +343,7 @@ int main(int argc, char ** argv) {
client.n_decoded = 0; client.n_decoded = 0;
client.i_batch = batch.n_tokens - 1; client.i_batch = batch.n_tokens - 1;
LOG_INF("\033[31mClient %3d, seq %4d, started decoding ...\033[0m\n", client.id, client.seq_id); LOG_INF("\033[31mClient %3d, seq %4d, junk = %4d, started decoding ...\033[0m\n", client.id, client.seq_id, n_junk_cur);
g_seq_id += 1; g_seq_id += 1;