Awesome-Efficient-LLM

Why we included this project

This is one of the more useful starting points we have found for anyone trying to make large language models cheaper to run, whether that means shrinking them, speeding up inference, or cutting memory use. The list is organized into focused sub-areas such as pruning, quantization, knowledge distillation, KV-cache compression, low-rank decomposition, and efficient fine-tuning, so you can jump straight to the technique you are investigating rather than wading through a flat pile of links. Each entry carries the paper, its code repository, and the venue it appeared at, which makes it practical for tracing a method from the write-up to a runnable implementation. It is maintained actively and accepts pull requests, so the collection stays current with recent work in the field. For teams scoping out which efficiency technique to adopt, this is a solid map of the options and the code behind them.