CVE-2026-71486 vLLM: Derender endpoints decode caller-supplied GenerateResponse token IDs without output bounds
An authenticated attacker can cause excessive resource consumption and oversized responses in vLLM by sending crafted token IDs to derender endpoints before limits are enforced.
- Exploited in the wild
- Not known
- Login needed
- Any signed-in user
- Attackable
- Over the internet
- Victim action needed
- Not needed
- Fix
- Available
What is CVE-2026-71486?
CVE-2026-71486 is a medium-severity denial of service vulnerability in Vllm Vllm, published on 17 Aug 2026. Any signed-in user can exploit it over the network, with no action from the victim.
Official 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.
Which versions are affected by CVE-2026-71486?
3 open source packages ship an affected version, including vllm. It affects 1 product.
| Package | Affected | Fixed in |
|---|---|---|
vllm PyPI | All versions | 0.26.0 |
vllm PyPI | All versions | 0.26.0 |
vllm PyPI | All versions | 0.26.0 |
Products include Vllm Vllm.
How do I fix CVE-2026-71486?
Upgrade vllm to 0.26.0 or later.
Upgrade, then put limits on request size, rate and nesting depth so one client cannot exhaust the service. More about Denial of service.
Is CVE-2026-71486 being exploited?
There is no known exploitation of CVE-2026-71486 in the wild and no public exploit code.
Other vulnerabilities in Vllm Vllm
- CVE-2025-32444 vLLM Vulnerable to Remote Code Execution via Mooncake Integration
- CVE-2026-22778 vLLM leaks a heap address when PIL throws an error
- CVE-2025-47277 vLLM Allows Remote Code Execution via PyNcclPipe Communication Service
- CVE-2024-11041 Remote Code Execution in vllm-project/vllm
- CVE-2026-48746 vLLM: OpenAI auth bypass
Sources
Also known as GHSA-8737-qx52-hjff, MINI-449j-7ph2-97pm, MINI-rhph-wf6r-cg97, MINI-wwj8-fv7x-7m55, PYSEC-2026-3936.