Hao AI Lab 用 vstack 对 vLLM 做智能体形式化验证,发现 17 个 bug
(1/7) Agentic Formal Verification on vLLM 🚀
Hao AI Lab 与 vLLM 项目合作,用 vstack 对 vLLM 做智能体形式化验证,已发现 17 个(仍在增加)bug,涵盖静默错误输出到服务器锁死等问题。团队公开了其中 5 个 bug 的成因与修复 PR,例如混合模型可从全零缓存块继续生成、队列上限下失败请求不归还槽位导致服务器拒绝所有新请求。相关修复正在向上游提交。
正文
(1/7) Agentic Formal Verification on vLLM 🚀
This week, we introduced vstack for agentic formal verification in AI infra; today, we're working with @vllm_project , putting vstack to test: although vLLM is already very mature, some bugs only surface in rare combinations that are hard for human reviewers to catch.
With vstack's agentic formal verification, we identified 17 (and growing!) bugs, from silent wrong output to server lockout. Below we share 5 of them, each with a fix PR ⬇️
🔌 vstack plugin: https://t.co/GZJebbjdpL 📋 Bugs Tracker: https://t.co/6f6WmtRUWU