SALMONN is an open-source project on GitHub with about 1,554 stars, most recently updated in 2026. SALMONN is a family of advanced multimodal large language models from ByteDance, developed with Tsinghua University researchers. The models give LLMs general hearing abilities across speech, audio events and music, and the family extends to audio-visual understanding, with work presented at ICLR and ICML 2024.
Published on GitHub under the Apache-2.0 license, SALMONN has attracted a community of about 1,554 stars, which places it among the more visible open-source AI projects in its niche. Development was still active as of 2026, based on the repository’s latest updates. Because the full source code is public, teams can audit how it works, adapt it to their own stack, and self-host it rather than depending on a closed hosted product.
The repository is tagged with topics including audio, audio-processing, audio-visual-understanding, bytedance, iclr2024, which gives a quick picture of where it fits in the open-source AI ecosystem. New users can explore the examples and documentation in the repository to understand the main workflows before integrating it into their own projects, and the public issue tracker shows what the maintainers are actively working on and what the community is asking for.
SALMONN is Apache-2.0 open source and free for research use under that license.
SALMONN is best suited for multimodal AI researchers working on speech, audio and video understanding who want strong open research models. Anyone evaluating it should start with the GitHub repository, where the documentation, examples and issue tracker show exactly what the current version can do.