Server: fix seed for multiple slots (#6835)

* Server: add tests for consistent results

* sampling: separate rng per sampling context

1 files changed, 0 insertions, 1 deletions

diff --git a/examples/main/main.cpp b/examples/main/main.cpp
index 1180734b..a74d4d9c 100644
--- a/examples/main/main.cpp
+++ b/examples/main/main.cpp
@@ -240,7 +240,6 @@ int main(int argc, char ** argv) {
                 return 1;
             }
             session_tokens.resize(n_token_count_out);
-            llama_set_rng_seed(ctx, params.seed);
             LOG_TEE("%s: loaded a session with prompt size of %d tokens\n", __func__, (int)session_tokens.size());
         }
     }