Machine-written summary
Summary written by Claude Haiku from the source advisory. Not reviewed by a person.
vLLM versions before 0.30.0 do not properly validate bad_words token indices against the model's generation output width in SamplingParams.update_from_tokenizer(). An attacker who can send requests to the service can supply out-of-bounds token indices, which corrupts logits memory and causes other concurrent requests to return incorrect tokens. Upgrade vLLM to version 0.30.0 or later.
Not analysed yet
We haven’t published our analysis of this Advisory yet. Once a person has reviewed it, this page will say what’s vulnerable, what has to be true for your app to be affected, and how to check.
Severity and exploitation
| Score | Rating | CVSS | Vector | Source |
|---|---|---|---|---|
| 2.3 | Low | 4.0 | CVSS:4.0/AV:N/AC:H/AT:P/PR:L/UI:N/VC:N/VI:L/VA:N/SC:N/SI:N/SA:N | CNA (VulnCheck) |
- CISA KEV Not listed.
- EPSS 0.0020, the 10th percentile, on 6 Nov 2026: FIRST’s estimate of the chance of exploitation in the next 30 days.
Timeline
- 31 Oct 2026, 14:09 UTC First seen by agentprime
- 31 Oct 2026, 14:09 UTC Confirmed by OSV
Sources
- OSV: PYSEC-2026-4000
- github.com/vllm-project/vllm/blob/98dff2a81d747d1dba01a47f939f48c3526d4206/vllm/sampling_params.py#L694-L753
- github.com/vllm-project/vllm/blob/98dff2a81d747d1dba01a47f939f48c3526d4206/vllm/v1/worker/gpu/sample/bad_words.py
- www.vulncheck.com/advisories/vllm-through-0.29.0-cross-request-logits-corruption-via-bad-words
- github.com/vllm-project/vllm/pull/48824
- github.com/vllm-project/vllm