vLLM streaming session out-of-bounds write vulnerability in shared cache
Automated summary · Based on collected summaries; not individually verified
Prepared by HotAI
What happened
vLLM has a silent out-of-bounds write issue: when streaming input, it only checks the length of each new chunk but fails to verify the total session length. This causes token data from excessively long sessions to overflow into shared prefix cache blocks, altering output for subsequent requests that reuse these blocks without triggering any errors. The issue can be fixed by restricting worker writes and discarding tokens that exceed limits. See PR vllm-project/vllm#67457.
Background
What is a token? Its relationship to text length, usage, and context budgets
A token is a discrete unit used to process text, potentially a character, part of a word, or punctuation depending on the tokenizer. It is not a fixed number of Chinese characters or English words. Budgets and usage often use tokens, but accounting differs by model and interface. Use the target model’s counting method and inspect reported usage.
What is a pull request? Diffs, reviews, checks, and merge state
A pull request proposes merging one branch into another and collects review, discussion, and checks. It differs from a commit and does not establish merge or deployment. AI-created PRs still need diff and test review under project rules. User availability depends on release and live verification.
Reading notes
The event summary comes from collected excerpts. Background passages quote existing HotAI articles to explain terms and do not establish new facts about this event. Check the original sources for product results, conditions and subsequent updates.
Original source
Sources and collection timeline
| Source | Report title | Original publication time | Collected by HotAI |
|---|---|---|---|
| x.com | vLLM streaming session out-of-bounds write vulnerability in shared cache | 2026-11-14 03:13 | 2026-11-14 03:37 |