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.
Affected products
vllm: vllm
References4
- github.com/vllm-project/vllm/security/advisories/GHSA-8737-qx52… ExploitVendor AdvisoryMitigation
- github.com/vllm-project/vllm/commit/8e61b646e2d157f9b93451fa048… Patch
- github.com/vllm-project/vllm/pull/47260 Issue TrackingPatch
- github.com/vllm-project/vllm/releases/tag/v0.26.0 Release Notes
Metrics
- CVSS 3.1
-
4.3
MEDIUM
Score source: CNA
CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:L - EPSS — exploit probability
- 0% percentile: 32 · as of 11 November 2026
- CWE
- CWE-400CWE-770
Could this be exploited in your environment?
A patch closes the known hole. A penetration test shows what an attacker can still reach — with proof.