I’ve been digging in the Langchain documentation (and type definitions) and even asked LLMs how I can upload or send an image to multi-modal LLMS, but I haven’t found anything.