add dry sampler (#513)

* add dry sampler

* use vocab instead of model in dry_init function

* fix compile error for build test

---------

Co-authored-by: firecoperana <firecoperana>

1 files changed, 1 insertions, 1 deletions

diff --git a/examples/parallel/parallel.cpp b/examples/parallel/parallel.cpp
index 621a1c95..ee614f84 100644
--- a/examples/parallel/parallel.cpp
+++ b/examples/parallel/parallel.cpp
@@ -161,7 +161,7 @@ int main(int argc, char ** argv) {
     for (size_t i = 0; i < clients.size(); ++i) {
         auto & client = clients[i];
         client.id = i;
-        client.ctx_sampling = llama_sampling_init(params.sparams);
+        client.ctx_sampling = llama_sampling_init(llama_get_model_vocab(model), params.sparams);
     }
 
     std::vector<llama_token> tokens_system;