CVE-2026-71486— vLLM: Derender endpoints decode caller-supplied GenerateResponse token IDs without output bounds
Possible ATT&CK Techniques 1AI
T1190 · Exploit Public-Facing ApplicationGet alerts for future matching vulnerabilitiesLog in to subscribe
I. Basic Information for CVE-2026-71486
Vulnerability Information
Have questions about the vulnerability? See if Shenlong's analysis helps!
View Shenlong Deep Dive ↗ Although we use advanced large model technology, its output may still contain inaccurate or outdated information.Shenlong tries to ensure data accuracy, but please verify and judge based on the actual situation.
Vulnerability Title
vLLM: Derender endpoints decode caller-supplied GenerateResponse token IDs without output bounds
Source: CVE Program / CVE List V5
Vulnerability 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.
Source: CVE Program / CVE List V5
CVSS Information
CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:L
Source: CVE Program / CVE List V5
Vulnerability Type
未加控制的资源消耗(资源穷尽)
Source: CVE Program / CVE List V5
Affected Products
| Vendor | Product | Affected Versions | CPE | Subscribe |
|---|---|---|---|---|
| vllm-project | vllm | < 0.26.0 | - |
II. Public POCs for CVE-2026-71486
| # | POC Description | Source Link | Shenlong Link |
|---|
AI-Generated POCPremium
No public POC found.
Login to generate AI POCIII. Intelligence Information for CVE-2026-71486
请登录查看更多情报信息。
Patches & Fixes for CVE-2026-71486 (2)
Vendor Advisories for CVE-2026-71486 (1)
Vendor Pages for CVE-2026-71486 (1)
- 🔗 Release v0.26.0 · vllm-project/vllm · GitHubx_refsource_MISC
IV. Related Vulnerabilities
Same product: vllm
- CVE-2026-73560
vLLM: SSRF + arbitrary local file read in MiMoV2OmniMultiMod - CVE-2026-73559
vLLM: Completion prompt lists fan out into unbounded engine - CVE-2026-73558
vLLM: Cross-User Data Leak Vulnerability - CVE-2026-73557
vLLM: Incomplete CVE-2025-62164 remediation can be bypassed - CVE-2026-73556
vLLM: ReDoS via structured_outputs.regex in the lm-format-en
Same vendor: vllm-project
- CVE-2026-73560
vLLM: SSRF + arbitrary local file read in MiMoV2OmniMultiMod - CVE-2026-73559
vLLM: Completion prompt lists fan out into unbounded engine - CVE-2026-73558
vLLM: Cross-User Data Leak Vulnerability - CVE-2026-73557
vLLM: Incomplete CVE-2025-62164 remediation can be bypassed - CVE-2026-73556
vLLM: ReDoS via structured_outputs.regex in the lm-format-en
Same weakness: CWE-400
- CVE-2026-65976
Deskflow: Clipboard receiver can accumulate data beyond Desk - CVE-2026-17639
Certain HP Smart Tank All in One – Potential Denial of Servi - CVE-2026-59902
Netty: Memory Exhaustion in SctpMessageCompletionHandler - CVE-2026-71491
sqlparse: Quadratic O(n²) DoS in group_comments - CVE-2026-64868
New API: Unauthenticated payment webhooks allow memory and d
V. Comments for CVE-2026-71486
No comments yet