Gemma 3n multimodal support plans?

[ai-fonsi]

https://huggingface.co/ggml-org/gemma-3n-E2B-it-GGUF says:

This version does not contain multimodal support. We are still working on adding multimodal.

Are there still any plans to add support? I haven't seen any activity towards multimodality in Gemma 3n in this repo. Support for this model specifically would be very useful as it's one of the only models available that can handle both audio and images at the same time, enabling novel use-cases.

[kha84]

Is there a single model supported by llama.cpp that can be fed with audio?

[3Simplex]

Today I tried the same model.

INFO:hf-to-gguf:Loading model: gemma-3n-E4B-it
WARNING:hf-to-gguf:Failed to load model config from C:\Users\Clayton\Llama.Cpp-Toolbox\llama.cpp\models\gemma-3n-E4B-it: The checkpoint you are trying to load has model type `gemma3n` but Transformers does not recognize this architecture. This could be because of an issue with the checkpoint, or because your version of Transformers is out of date.
WARNING:hf-to-gguf:Trying to load config.json instead
INFO:hf-to-gguf:Model architecture: Gemma3nForConditionalGeneration
ERROR:hf-to-gguf:Model Gemma3nForConditionalGeneration is not supported

[AyushWalekar]

Here after 3 months, still no updates. Really looking forward to this.

[simrnsingh]

Thought this would be an appropriate place for discussion about Audio support for Gemma3n.
I wanted to confer, how bullish are you on adding audio support? @ngxson
I see you have quite a lot of work already done on the audio encoder, but I would be happy to contribute if you are not actively working on it.
I agree with comments above; gemma3n is quite unique in its support for both vision and audio especially considering its params count. Please let me know if you think this is worth the effort!