Pocket-Sized Multimodal AI for content understanding and generation across multilingual texts, images, and 🔜 video, up to 5x faster than OpenAI CLIP and LLaVA 🖼️ & 🖋️
The current @unum-cloud/uform package uses onnxruntime-node. The WASM-based in-browser alternatives should be easy to swap in. It would be great to provide the user with a knob to select the backend or detect it automatically 🤗
The current
@unum-cloud/uformpackage usesonnxruntime-node. The WASM-based in-browser alternatives should be easy to swap in. It would be great to provide the user with a knob to select the backend or detect it automatically 🤗