Quesma has released an open, curated list containing 218 entries that track where tokens are spent in LLM and agent workflows. The resource covers monitoring, semantic caching, cheap local models, context engineering, and cost governance.
Each entry includes a citation and date, with the list deliberately excluding claims that lack supporting numbers or methods. It aims to distinguish between tokens that are wasted versus those well spent.
The maintainers invite corrections for inaccuracies and welcome suggestions for new entries, which undergo the same review process as existing content.