kohjingyu/gill
A multimodal LLM that processes interleaved image-and-text inputs to generate text, retrieve images, and synthesize images.
Not currently ranked — collecting fresh signals.
star history
GILL (Generating Images with Large Language Models) is a NeurIPS 2023 research project that extends an LLM with vision capabilities. It enables the model to process arbitrarily interleaved image-and-text inputs and produce outputs including text responses, retrieved images from a large collection, and newly generated images. The model bridges large language models with image generation and retrieval using learned projection layers.
Frequently asked
- What is kohjingyu/gill?
- A multimodal LLM that processes interleaved image-and-text inputs to generate text, retrieve images, and synthesize images.
- Is gill open source?
- Yes — kohjingyu/gill is open source, released under the Apache-2.0 license.
- What language is gill written in?
- kohjingyu/gill is primarily written in Jupyter Notebook.
- How popular is gill?
- kohjingyu/gill has 472 stars on GitHub.
- Where can I find gill?
- kohjingyu/gill is on GitHub at https://github.com/kohjingyu/gill.