Shanghai Innovation Institute (SII) · Fudan University · MOSI.AI
Open, collaborative research on Large Language Models and Multimodal Foundation Models.
OpenMOSS is a research group led by Prof. Xipeng Qiu, hosted at the Shanghai Innovation Institute (SII) and working in close collaboration with Fudan University and MOSI.AI. We conduct cutting-edge research across the full LLM stack — from model architecture and training to evaluation, interpretability, and real-world applications — with a strong commitment to open and reproducible science.
| Direction | Flagship Repositories |
|---|---|
| 🧠 Foundation LLMs | MOSS |
| 👁️ Vision & Video | MOSS-VL · MOSS-Video-Preview · MOVA |
| 🌐 Omni LLMs | AnyGPT |
| 🗣️ Speech / TTS | MOSS-TTS · MOSS-TTS-Nano · MOSS-TTSD · MOSS-Audio-Tokenizer |
| 🗣️ Speech / Transcribe | MOSS-Transcribe-Diarize |
| 🗣️ Speech / Interaction | MOSS-Speech · SpeechGPT-2.0-preview |
| 🎵 Audio & Music | MOSS-Audio · MOSS-Music |
| 🤖 Embodied AI & Robotics | RoboOmni · FRoM-W1 |
| 🔍 Interpretability | Llamascopium (formerly Language-Model-SAEs) · Lorsa |
| 📊 Benchmarks & Evaluation | VLABench · GAOKAO-MM · Say-I-Dont-Know |
| ⚡ Efficiency & Long Context | CoLLiE · LongLLaDA · rope_pp |
| 📚 Survey | Awesome-WAM · Thus-Spake-Long-Context-LLM |
- MOSS-TTS-Nano — 0.1B-param multilingual TTS that runs directly on CPU, no GPU required · ~3.9k★
- MOSS-TTS — Expressive speech & sound-generation family: long-form, multi-speaker, voice/character design, and streaming TTS · ~3.7k★
- MOVA — Scalable and synchronized video–audio generation · ~1.1k★
- Awesome-WAM — Continuously updated reading list & resources for World Action Models in embodied AI · ~1.1k★
- MOSS-Audio — Unified audio-understanding foundation model spanning speech, sound, music, captioning, QA & reasoning
- MOSS-VL — Core multimodal vision-understanding series with the full training stack open-sourced
- MOSS-Transcribe-Diarize — Speech transcription with speaker diarization (who spoke when) for multi-speaker, long-form audio
See the pinned repositories for quick access, or browse all 50+ repositories.
We welcome researchers, students, and collaborators who share our vision. For PhD/intern openings, research collaborations, or general inquiries, please reach us at openmoss@sii.edu.cn.
The Shanghai Innovation Institute (SII) is dedicated to fostering innovation in education and research in the field of artificial intelligence.
