Vulnerability PYSEC-2026-4000

Medium Risk
MEDIUM RISK
CVSS Score: 4.3
Score Range: 4.0–6.9
Medium severity vulnerabilities (CVSS 4.0–6.9). Important issues that meaningfully reduce security confidence.
9 days ago
September 19, 2026 at 11:17 PM UTC
No summary available
0.0.1 - 0.29.0
0.0.1 - 0.29.0

Details

vLLM through 0.29.0 fails to properly validate bad_words token indices against the model's generation output width in SamplingParams.update_from_tokenizer(). Attackers can supply out-of-bounds token indices that corrupt logits memory of concurrent requests, causing different in-flight HTTP requests to return incorrect tokens.

Impacted packages

Timeline

Published
9 days ago
September 19, 2026 at 11:17 PM UTC
Fixed (0.30.0)
7 days ago
September 22, 2026 at 05:23 AM UTC
Last Modified
2 hours ago
September 29, 2026 at 09:00 AM UTC