CVE-2026-93989: vllm · Improper Validation of Array Index
Affects vllmSource CVE Database V5
Not known to be exploited · CVSS 2.3 Low.
Data as of 2 Nov 2026, 22:58 UTC
- EPSS
- 0.2% chance of exploitation in 30 days – higher than 10% of CVEsExploit Prediction Scoring System: the likelihood attackers will use this flaw soon.
CVSS:4.0/AV:N/AC:H/AT:P/PR:L/UI:N/VC:N/VI:L/VA:N/SC:N/SI:N/SA:NExploitation evidence
Disclosed: evidence found.
CVE published 19 Sep 2026
Detectable: no evidence on record.
No public nuclei template indexed
Weaponized: no evidence on record.
No public exploit code indexed in Metasploit or Exploit-DB
Exploited in the wildExploited: no evidence on record.
Not in CISA KEV and no source reports exploitation
Observed by OffSeq MirageSeen by Mirage: no evidence on record.
No attack traffic recorded by OffSeq Mirage sensors
A missing step is not evidence of safety. Sources: cve.org (opens in a new tab), the Metasploit, Exploit-DB and nuclei-templates index as of 2 Nov 2026.
What to do
Affected software
| Product | Affected | Fixed in |
|---|---|---|
| vllm-projectvllm | Affected
| Fixed in– |
GitHubvllmpkg:github/vllmOther threats | Affected
| Fixed in– |
Check your own hosts for these packages with threat-finder (opens in a new tab), our free open-source scanner.
AI-suggested
Until a fix is available, avoid supplying untrusted or malformed bad_words token indices in requests to mitigate risk.
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.
References
- github.com/vllm-project/vllm/pull/48824 (opens in a new tab)
- github.com/vllm-project/vllm/blob/98dff2a81d747d1dba01a47f939f48c3526d4206/vllm/v1/worker/gpu/sample/bad_words.py (opens in a new tab)
- github.com/vllm-project/vllm/blob/98dff2a81d747d1dba01a47f939f48c3526d4206/vllm/sampling_params.py (opens in a new tab)
- github.com/vllm-project/vllm (opens in a new tab)
- vulncheck.com/advisories/vllm-through-0.29.0-cross-request-logits-corruption-via-bad-words (opens in a new tab)