FlexGen

★ 1 updated 3y ago ⑂ fork

Throughput-oriented systems for large language models on commodity GPUs.

No plain-English explanation yet — one is being written right now. Check back in a minute.