We ran a set of popular open-source models against our IDOR benchmark, the same dataset and the same prompt we’ve used to evaluate frontier coding agents. The result surprised us: GLM 5.2, an open-weight model from Zhipu AI, scored a … Read more
Image by Annie Ruygt This bit by Geoffrey Huntley is super interesting to me and, despite calling out that LLM-driven development agents like Cursor have something like a 40% success rate at actually building anything that passes acceptance criteria, makes … Read more
https://semgrep.dev/products/semgrep-app Hi! I work on Semgrep, an open-source project (discussed on HN previously [0][1)]. We’re one of those companies that maintain an OSS tool and a web app, and then monetize by selling enterprise features on said web app. Our … Read more