A pull request (#31028) has been submitted to the llama.cpp repository proposing to merge tensors during the model conversion process, which may require users to reconvert existing GGUF files. The change aims to optimize tensor handling in the Qwen model conversion pipeline. Users are advised to reconvert their GGUF models to benefit from this update.