Hao AI Lab 用 vstack 对 vLLM 做智能体形式化验证,发现 17 个以上 bug
(7/7) Our fixes are going upstream now, and we're continuing to work with the vLLM team and the open-source community to bring more verifiable AI infra.
Hao AI Lab 用 vstack 对 vLLM 做智能体形式化验证,已发现 17 个(还在增加)bug,并公开其中 5 个及对应修复 PR。问题包括混合模型从全零缓存块继续生成、队列上限下少量失败请求导致服务器持续返回 busy、长流式会话越界写入共享缓存块、LoRA 适配器原地重载后新请求仍用旧权重,以及两台服务器共享外部 KV 缓存时不同图片命中同一缓存键。相关修复正在向上游提交,团队表示将继续与 vLLM 团队和开源社区合作。
正文
(7/7) Our fixes are going upstream now, and we're continuing to work with the vLLM team and the open-source community to bring more verifiable AI infra.
Huge thanks to the @vllm_project team for building such a robust engine and for their valuable feedback!
vLLM verification work by @AlexZhang152005 and @Junda_Su08 , with special thanks to @Yuxuan_Zhang13 , @yisun0618 , @Junda_Chen_ and @FuYichao123 for building this together at @haoailab !