ggml-cuda : use graph allocator (#2684)

use a different function for no_alloc to avoid breaking backwards compat, fixes lora

remove 512 n_batch limit

fixed 2048 batch size

cleanup

Co-authored-by: Johannes Gäßler <johannesg@5d6.de>

1 files changed, 0 insertions, 1 deletions

diff --git a/common/common.cpp b/common/common.cpp
index d7e1a572..1623ba21 100644
--- a/common/common.cpp
+++ b/common/common.cpp
@@ -289,7 +289,6 @@ bool gpt_params_parse(int argc, char ** argv, gpt_params & params) {
                 break;
             }
             params.n_batch = std::stoi(argv[i]);
-            params.n_batch = std::min(512, params.n_batch);
         } else if (arg == "--keep") {
             if (++i >= argc) {
                 invalid_param = true;