The word MOSS in giant cream-colored letters on a yellow backdrop, with an elephant in a cowboy hat, a felt doll holding a red rose, and a python wearing sunglasses — by MOSS-Image

Yang Gao 高扬

Member of Technical Staff
Mosi Intelligence (MOSI), Shanghai, China

Email: gaoy2140@gmail.com
WeChat: garythevoid

About

I am a Member of Technical Staff at Mosi Intelligence (MOSI), focusing on performance optimization for large-scale training and inference systems. As part of the OpenMOSS team, I work on multimodal foundation models across image, video, audio and speech, spanning both understanding and generation.

Previously, I worked on large language model pre-training at InternLM (书生·浦语) in Shanghai AI Laboratory. I received my M.S. (2022) and B.S. (2019) degrees from the School of Computer Science and Technology at Northwestern Polytechnical University (NWPU), advised by Prof. Peng Wang.

My research and engineering interests focus on:

Large-Scale Training Systems
High-Performance Inference Systems
Multimodal Understanding and Generation

We are hiring! We are looking for engineers and researchers interested in AI infrastructure, large-scale training/inference systems, and multimodal foundation models. Feel free to reach out if you are interested in joining us.

Open-Source Collaborations

OpenMOSS maintains deep, official collaborations with the SGLang and vLLM teams, bringing first-class serving support for MOSS models to their omni-modal runtimes.

SGLang's high-performance serving runtime for interactive, omni-modal systems.

Serves MOSS-TTS, MOSS-TTS-Local and MOSS-Transcribe-Diarize

GitHub Docs

vLLM's framework for easy, fast and cheap omni-modality model serving.

Serves the MOSS-TTS family and MOSS-Image

GitHub Docs

Featured Posts

SGLang Blog · Oct 20, 2026 Unified Runtime for Full-Duplex Interactive Intelligence with SGLang Omni
One runtime for native full-duplex models, turn-based multimodal models and cascaded speech systems, built on composable graphs and persistent sessions. On a single H200, an ASR → LLM → TTS cascade reaches 0.17 s mean stop latency on interruption with 16 concurrent sessions.
LMSYS Blog · Jun 17, 2026 MOSS-TTS Local Transformer v1.5 on SGLang-Omni: Serving Native-Streaming 48 kHz Speech
Serving the 48 kHz stereo MOSS-TTS-Local-Transformer-v1.5 as a three-stage pipeline (preprocessing → AR engine → streaming vocoder), with reference-audio caching and CUDA Graph capture for native streaming.

Selected Publications

2026

MOSS-Image Technical Report

A 5B single-stream DiT trained from scratch for bilingual long-text rendering and precise layout control.

OpenMOSS Team

arXiv Paper Code Hugging Face Project
MOSS-VL Technical Report

An open-weight 11B model series for long-form and real-time video understanding.

OpenMOSS Team

arXiv Paper Code Hugging Face Project
MOSS-Audio Technical Report

Unified understanding of speech, environmental sound and music through captioning, question answering and reasoning.

OpenMOSS Team

arXiv Paper Code Hugging Face Project
MOSS-TTS Technical Report

An open model family for long-form speech, dialogue synthesis, voice design, sound effects and real-time streaming TTS.

OpenMOSS Team

arXiv Paper Code Hugging Face Project
MOVA: Towards Scalable and Synchronized Video-Audio Generation

OpenMOSS Team

arXiv Paper Code Hugging Face Project
MOSS-Speech: Towards True Speech-to-Speech Models Without Text Guidance

Xingjian Zhao, Zhe Xu, Qinyuan Cheng, Zhaoye Fei, Luozhijie Jin, Yang Wang, Hanfu Chen, Yaozhou Jiang, Qinghui Gao, Ke Chen, Ruixiao Li, Mingshu Chen, Ruiming Wang, Wenbo Zhang, Yiyang Zhang, Donghua Yu, Yang Gao, Xiaogui Yang, Yitian Gong, Yuanfan Xu, Yaqian Zhou, Xuanjing Huang, Xipeng Qiu

ICLR 2026 Paper arXiv Code Hugging Face

2025

Intern-S1: A Scientific Multimodal Foundation Model

InternLM Team

arXiv Paper Code Hugging Face

2024

InternLM-XComposer-2.5: A Versatile Large Vision Language Model Supporting Long-Contextual Input and Output

Pan Zhang, Xiaoyi Dong, Yuhang Zang, Yuhang Cao, Rui Qian, Lin Chen, Qipeng Guo, Haodong Duan, Bin Wang, Linke Ouyang, Songyang Zhang, Wenwei Zhang, Yining Li, Yang Gao, Peng Sun, Xinyue Zhang, Wei Li, Jingwen Li, Wenhai Wang, Hang Yan, Conghui He, Xingcheng Zhang, Kai Chen, Jifeng Dai, Yu Qiao, Dahua Lin, Jiaqi Wang

arXiv Paper Code Hugging Face
Lins: Reducing Communication Overhead of ZeRO for Efficient LLM Training

Qiaoling Chen, Qinghao Hu, Guoteng Wang, Yingtong Xiong, Ting Huang, Xun Chen, Yang Gao, Hang Yan, Yonggang Wen, Tianwei Zhang, Peng Sun

IWQoS 2024 Paper
InternLM-XComposer2-4KHD: A Pioneering Large Vision-Language Model Handling Resolutions from 336 Pixels to 4K HD

Xiaoyi Dong, Pan Zhang, Yuhang Zang, Yuhang Cao, Bin Wang, Linke Ouyang, Songyang Zhang, Haodong Duan, Wenwei Zhang, Yining Li, Hang Yan, Yang Gao, Zhe Chen, Xinyue Zhang, Wei Li, Jingwen Li, Wenhai Wang, Kai Chen, Conghui He, Xingcheng Zhang, Jifeng Dai, Yu Qiao, Dahua Lin, Jiaqi Wang

NeurIPS 2024 Paper Code Hugging Face
InternLM2 Technical Report

InternLM Team

arXiv Paper Code Hugging Face

2023

InternLM: A Multilingual Language Model with Progressively Enhanced Capabilities

InternLM Team

Tech Report Paper Code Hugging Face

2020

NAS-FCOS: Fast Neural Architecture Search for Object Detection

Ning Wang*, Yang Gao*, Hao Chen*, Peng Wang, Zhi Tian, Chunhua Shen, Yanning Zhang (* equal contribution)

CVPR 2020 Paper Code

Experience

Industry

2025.07 – PresentMosi Intelligence (MOSI)Member of Technical Staff
2022.06 – 2025.06Shanghai AI LaboratoryResearch Engineer, Large Language Model Pre-training
2021.05 – 2022.03SenseTime ResearchEngineering Intern, Automated Machine Learning
2020.12 – 2021.05Baidu ResearchResearch Intern, Lightweight Object Detection

Academia

2019.09 – 2022.04Northwestern Polytechnical UniversityM.S. in Computer Science (Postgraduate Recommendation)
2015.09 – 2019.06Northwestern Polytechnical UniversityB.S. in Computer Science

Honors & Awards

2022Outstanding Graduate of Master Students (Top 5%), NWPU
2020, 2021, 2022Excellent First-class Scholarship for Master Students (three consecutive years), NWPU
2021Intern Future Star (Top 5%), SenseTime
2019Outstanding Graduate of Bachelor Students (Top 10%), NWPU
2017, 2018Excellent First-class Scholarship for Bachelor Students, NWPU