CVE-2026-71486: Vllm

Medium severity, CVSS 4.3. EPSS: 0.4% chance of exploitation in the next 30 days.

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.

Affected products

• Vllm Vllm: before 0.26.0 (fixed in 0.26.0)

Published 2026-08-17. Last modified 2026-10-02.