Nougat (Neural Optical Understanding for Academic Documents) is a vision encoder-decoder model developed by Meta. The base model is trained to read scientific PDFs and output structured Markdown that preserves equations, tables, and formatting. It is particularly useful for digitizing academic literature and making it accessible to language models or search systems.
In the Foundation models & chat space, Nougat Base takes a focused approach. It focuses on converting academic PDF documents containing complex math and tables into machine-readable Markdown or LaTeX. It is built as an open-source project for researchers. Nougat Base is open source under the MIT license. It runs on the web and API.
Meta AI builds and maintains Nougat Base, and the product first shipped in 2023. Development happens publicly on GitHub with 10k stars. Key capabilities include PDF Parsing, Math OCR, and Vision Encoder-Decoder.
Latest indexed changes and source events
facebook/nougat-base verified by the PulseGate indexer
Other apps tracked under the same category.