Search code, repositories, users, issues, pull requests...
The YAZKB (Yet Another Zero Knowledge Benchmark) project aims to offer an end-to-end, cutting-edge, and production-focused perspective by providing a unified, open and contemporary performance evaluation of the key zero-knowledge libraries, frameworks, and virtual machines.
Based on the results of our benchmarks, we provide evaluative reports with a clear stance. The purpose of a benchmark is to guide developers in understanding the latest advancements in the field of ZK performance engineering, thereby identifying appropriate technological paths. We encourage open discussion rather than no-commented codes.
From a production standpoint, the unit of system should be benched end-to-end. Our benchmarks are based on the perspective of user's usage, rather than the system's components (such as front-end or back-end). Furthermore, one advantage of this approach is that all supported zero-knowledge libraries, frameworks, zkvms and even zkevms can be benched together as long as they fulfill the same specific tasks.
The author is an expert who has worked in the fields of DB, AI, and languages for a long time. He believes that his past experience can provide ZK community with some unique perspectives on the evolution of the ZK framework.
