CVE catalog

CVE-2026-71486

vLLM is an inference and serving engine for large language models. Prior to 0.26.0, the /v1/completions/derender and /v1/chat/completions/derender endpoints accept caller-supplied GenerateResponse obj

[CVE catalog]

Severity
Medium · CVSS 4.3CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:L
Published
Updated
Source
[NVD] · [CVE record]
EPSS
0.37% · Oct 25, 2026

vLLM is an inference and serving engine for large language models. Prior to 0.26.0, the /v1/completions/derender and /v1/chat/completions/derender endpoints accept caller-supplied GenerateResponse objects whose generate_responses, choices, token_ids, prompt_logprobs, logprobs.content, top_logprobs, and routed_experts structures are processed by OnlineDerenderer and tokenizer.decode before max_model_len, max_tokens, max_num_seqs, or response-size limits are enforced, allowing an authenticated API client to consume excessive CPU and memory and produce oversized responses. This issue is fixed in version 0.26.0.

Vulnerable and fixed packages

Distribution-reported package versions and fixes. These do not depend on matching against the OpenFactory package catalog.

DistributionPackageReleaseVersionStatus
PipvllmNot reported0.26.0fixed
CpevllmNot reported0.26.0fixed

Fixed-version evidence

DistributionPackageTypeReleaseFixed versionUrgencyOriginBugs
PipvllmpipNot reported0.26.0Not reportedGHSA-8737-QX52-HJFFNot reported
CpevllmcpeNot reported0.26.0Not reportedNVD-CVE-2026-71486Not reported

OpenFactory catalog packages

No matching packages in this catalog.

Observed:

References