README.md

FastGen [Paper]

We analyze the architecture and potential performance bottlenecks of the generative models andpropose novel techniques to improve generative model inference latency and scalabilty.

Citation

@article{ryu2023fast,
  title={A Fast and Scalable Generative Model Inference on Distributed Multi-GPU Environment},
  author={Ryu, Junyeol and Kim, Jinpyo and Lee, Jaejin},
  journal={한국정보과학회 학술발표논문집},
  pages={28--30},
  year={2023}
}