vLLM v0.31.0 Ships CRIU Snapshots and Preload to Kill Cold Starts

vLLM v0.31.0 landed with 717 commits from 307 contributors, adding a preload daemon that keeps quantized weights resident in GPU memory and CRIU-based snapshots that turn cold starts into process-level restores.

2026-10-07 ~ 2026-10-07 · 3 related posts

• vLLM v0.31.0 ships 717 commits: vllm preload keeps quantized weights in GPU memory across restarts — lmoroney · 2026-10-07
• vLLM v0.31.0 uses CRIU snapshots to restore live TP1 engines without reloading weights — vllm_project · 2026-10-07

1 near-duplicate retellings: AccBalanced