1  ---
2  features:
3    - |
4      Added partial support for the ``image-text-to-text`` task in ``HuggingFaceLocalChatGenerator``.
5  
6      This allows the use of multimodal models like Qwen 3.5 or Ministral with text-only inputs.
7      Complete multimodal support via Hugging Face Transformers might be addressed in the future.