low · CVSS v3 3.1 · CVSS v4 2.3
CVE-2026-93989
vLLM through 0.29.0 fails to properly validate bad_words token indices against the model's generation output width in Sa…
Description
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.
Scores
- Severity
- low
- CVSS v2
- 2.1
- CVSS v3
- 3.1
- CVSS v4
- 2.3
- EPSS
- —