Operator programming language represented by Triton, is reshaping the new paradigm of AI chip programming. With Python-like syntax, highly efficient code that rivals CUDA kernels, and a natural affinity with the PyTorch ecosystem, Triton demonstrates extreme developer-friendliness, high performance, and immense potential. Even developers without expert-level GPU knowledge can step into the realm of AI chip programming.

On January 9, 2026, TritonNext Conference will be held in Beijing, jointly organized by FlagOS open-source community and Beijing Academy of Artificial Intelligence. Named "Triton Next," the conference aims to explore the future trends of next-generation operator programming languages represented by Triton, the evolving directions of Triton-like language compilers, and their broad application scenarios.

As the first conference in China dedicated to Triton technology and its ecosystem, we welcome global AI researchers, high-performance computing engineers, compiler experts, programming language enthusiasts, and all developers looking to overcome the complexities of AI chip programming to define the new future of AI chip software!

About FlagOS Community  

website: https://flagos.io

To enable the scalable use of heterogeneous AI hardware with diverse architectures, Beijing Academy of Artificial Intelligence (BAAI), in collaboration with Peking University, Tsinghua University, the Chinese Academy of Sciences (CAS), and dozens of industry partners across AI hardware, system software, algorithms, and research institutions, launched the FlagOS community in 2022.

FlagOS is a fully open-source system software stack designed to unify the model–system–chip layers and foster an open and collaborative ecosystem. It enables a “develop once, run anywhere” workflow across diverse AI accelerators, unlocking hardware performance, eliminating fragmentation among vendor-specific software stacks, and substantially lowering the cost of porting and maintaining AI workloads. The mission of FlagOS is to enable open AI computation and lower the barrier to AI usage and innovation worldwide.