vLLM v0.25.1 发布：修复启动阻塞与混合精度量化问题

AI 摘要

vLLM 发布 v0.25.1 补丁版本，包含两项关键修复：一是避免在系统缺少 FFmpeg 时因 TorchCodec 导入错误而阻塞模型启动；二是修复混合精度下 allreduce 与 RMSNorm 量化融合可能导致的输出损坏问题。该版本由两位贡献者提交，其中一位为首次贡献。

正文节选

# vLLM v0.25.1 ## Highlights This release features 2 commits from 2 contributors (1 new)! v0.25.1 is a patch release containing two targeted bug fixes on top of v0.25.0. ### Bug Fixes * **Avoid blocking model launching when no system FFmpeg is available for TorchCodec** (#47888). Previously `import torchcodec` raised a `RuntimeError` at import time when system FFmpeg was missing, which blocked startup (e.g. `vllm serve Qwen/Qwen3-VL-2B-Instruct`) even when TorchCodec was not in use.

发布时间：2026-07-14 16:51

抓取时间：2026-08-02 00:23

来源机构：vLLM