CVE-2026-71486

Description

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.

Traduire en français

Produits concernés

• vllm-project vllm
• Vllm vllm

Scores CVSS

Faiblesses (CWE)

• CWE-400
• CWE-770

Références

• https://github.com/vllm-project/vllm/commit/8e61b646e2d157f9b93451fa048f9c8530c8a67b Patch
• https://github.com/vllm-project/vllm/pull/47260 Issue TrackingPatch
• https://github.com/vllm-project/vllm/releases/tag/v0.26.0 Release Notes
• https://github.com/vllm-project/vllm/security/advisories/GHSA-8737-qx52-hjff ExploitVendor AdvisoryMitigation

Ailleurs

• NVD
• CVE.org
• CIRCL Vulnerability-Lookup
• CERT-FR
• GitHub Advisories