vLLM: Derender endpoints decode caller-supplied GenerateResponse token IDs without output bounds
cve-program · CVE-2026-71486
PublishedvLLM 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.
Read the publisher's advisoryAffected versions, risk and source history
Related advisories and sources
Publishers link these advisories through the identifiers shown below. A shared identifier or date alone does not tell you whether your software is affected or whether the advisories describe the same cause.
CVE-2026-71486 · CVE overview and exploitation status
cve-program · CVE-2026-71486
Related changes
Linked by CVE identifierNo related changes are available here.