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

batched : add bench tool (#3545)

Browse source 
* batched : add bench tool

* batched : minor fix table

* batched-bench : add readme + n_kv_max is now configurable

* batched-bench : init warm-up batch

* batched-bench : pass custom set of PP, TG and PL

* batched-bench : add mmq CLI arg
This commit is contained in:
Georgi Gerganov 2023-10-11 21:25:33 +03:00 • committed by GitHub
parent 24ba3d829e
commit 8c70a5ff25
No known key found for this signature in database
GPG key ID: 4AEE18F83AFDEB23
7 changed files with 321 additions and 3 deletions
Download patch file Download diff file Expand all files Collapse all files

2
examples/batched/batched.cpp
View file

@ -66,7 +66,7 @@ int main(int argc, char ** argv) {
ctx_params.seed = 1234;
ctx_params.n_ctx = n_kv_req;
ctx_params.n_batch = std::max(n_len, n_parallel);
ctx_params.n_threads = params.n_threads;
ctx_params.n_threads = params.n_threads;
ctx_params.n_threads_batch = params.n_threads_batch == -1 ? params.n_threads : params.n_threads_batch;
llama_context * ctx = llama_new_context_with_model(model, ctx_params);