feat: Image/multimodality support

[martinstoyanov-brex]

Hey folks, just to confirm the repo currently does not support image input to LLMs (output text)? Is this a limitation of the local models or just need to be added in? Thank you.

[guillermovahi]

Same question here

[pacehutt]

i would say, try https://www.npmjs.com/package/llama.rn/v/0.2.0-rc.1