Cheng Li
Source-listed: Member of Technical Staff · Black Forest Labs
Record snapshot: Sep 3, 2026. Coverage and affiliations may be incomplete or historical.
Selected work
5- DeepSpeed Inference: Enabling Efficient Inference of Transformer Models at Unprecedented ScaleSaved credit: Contributor · 2022Explore people connected to this work →
- DeepSpeedSaved credit: ContributorExplore people connected to this work →
- DeepSpeed Data Efficiency: Improving Deep Learning Model Quality and Training Efficiency via Efficient Data Sampling and RoutingSaved credit: Contributor · 2022Explore people connected to this work →
- Understanding INT4 Quantization for Transformer Models: Latency Speedup, Composability, and Failure CasesSaved credit: Contributor · 2023Explore people connected to this work →
- LLM-Analysis: Latency and Memory Analysis of Transformer Models for Training and InferenceSaved credit: Creator · 2023Explore people connected to this work →