Large-scale models are revolutionizing deep learning and AI research, driving major improvements in language understanding, generating creative texts, multi-lingual translation and many more. But despite their remarkable capabilities, the models’ large size creates latency and cost constraints that hinder the deployment of applications on top of them. In particular, increased inference time and memory consumption […]
ChatGPT只是前菜，2023要来更大的！ - 墨天轮
PDF] DeepSpeed- Inference: Enabling Efficient Inference of Transformer Models at Unprecedented Scale
Michel LAPLANE (@MichelLAPLANE) / X
This AI newsletter is all you need #6
ZeroQuant与SmoothQuant量化总结-CSDN博客
ChatGPT只是前菜，2023要来更大的！ - 墨天轮
DeepSpeed ZeRO++: A leap in speed for LLM and chat model training with 4X less communication - Microsoft Research
PDF] DeepSpeed- Inference: Enabling Efficient Inference of Transformer Models at Unprecedented Scale
PDF] DeepSpeed- Inference: Enabling Efficient Inference of Transformer Models at Unprecedented Scale
如何评价微软开源的分布式训练框架deepspeed？ - 菩提树的回答- 知乎
GitHub - microsoft/DeepSpeed: DeepSpeed is a deep learning optimization library that makes distributed training and inference easy, efficient, and effective.
