talk-llama : sync llama.cpp

This commit is contained in:
Georgi Gerganov
2026-05-14 21:26:48 +03:00
parent 4730e76552
commit 54ecc9dba4
144 changed files with 11297 additions and 8333 deletions
+6 -1
View File
@@ -65,8 +65,13 @@ static ggml_tensor * ggml_mul_mat_aux(
ggml_tensor * res;
res = ggml_reshape_2d(ctx, cur, n, ggml_nelements(cur)/n);
if (!ggml_is_contiguous(cur)) {
res = ggml_cont_2d (ctx, cur, n, ggml_nelements(cur)/n);
} else {
res = ggml_reshape_2d(ctx, cur, n, ggml_nelements(cur)/n);
}
res = ggml_mul_mat (ctx, rot, res);
ggml_mul_mat_set_hint(res, GGML_HINT_SRC0_IS_HADAMARD);
res = ggml_reshape_4d(ctx, res, cur->ne[0], cur->ne[1], cur->ne[2], cur->ne[3]);
return res;