diff --git a/ggml-cuda.cu b/ggml-cuda.cu
index fe27bbac4..20dfd7561 100644
--- a/ggml-cuda.cu
+++ b/ggml-cuda.cu
@@ -7044,7 +7044,6 @@ inline void ggml_cuda_op_rope(
         GGML_ASSERT(false);
         rope_glm_f32_cuda(src0_dd, dst_dd, ne00, nrows, pos, freq_scale, ne01, freq_base, n_ctx, main_stream);
     } else if (is_neox) {
-        GGML_ASSERT(ne00 == n_dims && "ne00 != n_dims is not implemented for CUDA yet");
         if (src0->type == GGML_TYPE_F32) {
             rope_neox_cuda(
                 (const float *)src0_dd, (float *)dst_dd, ne00, n_dims, nrows, pos, freq_scale, ne01, freq_base, ext_factor,