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.
CVE-2026-71486
“—” beside the score: it is equal to the CVSS score beside it.
This medium-severity CVE scores 4.3 under a secondary CVSS source (NVD's own analysis pending). EPSS exploit probability: 0.4% (29th percentile of EPSS-scored CVEs). No GitHub Security Advisory CVSS is on record for this CVE.
- Lower severity and no public exploit yet
A fix is available — apply it.
- CVSS v3
- 4.3
- EG Score
- 4.3MEDIUMmedium confidence
- EG Risk
- 19EG Risk 19/100
EG Risk is EchelonGraph's 0–100 priority score: it fuses intrinsic severity with real-world exploitation and automatability so you can rank equal-severity CVEs and fix the most dangerous first. Higher = act sooner. Distinct from the 0–10 EG Score (severity).
How it’s computed (value × EchelonGraph’s weight)Severity43% × 45%The EG Score, 4.3 of 10.Exploitation0.4% × 40%FIRST EPSS: the 30-day exploitation probability when EchelonGraph scored this CVE.Automatability0% × 15%CISA's SSVC Automatable value: no (Vulnrichment).SSVC (CISA's decision table, applied by EchelonGraph): Track at every mission impact level.Exploitation none (CISA Vulnrichment) · Automatable no (CISA Vulnrichment) · Technical impact partial (CISA Vulnrichment). All three are CISA's SSVC values (Vulnrichment). EchelonGraph applied CISA's decision table to them; CISA publishes SSVC inputs and the table, not a decision for each CVE.Action: A fix is available. Apply it within your standard update timelines. - EPSS PROB
- 0.4%
- EPSS %ILE
- 29th
- KEV
- Not listed
SSVC (CISA's decision table, applied by EchelonGraph)Track at every mission impact level.
A fix is available. Apply it within your standard update timelines.
Exploitation none (CISA Vulnrichment) · Automatable no (CISA Vulnrichment) · Technical impact partial (CISA Vulnrichment). All three are CISA's SSVC values (Vulnrichment). EchelonGraph applied CISA's decision table to them; CISA publishes SSVC inputs and the table, not a decision for each CVE. Mission impact is CISA's Mission & Well-being decision point, and only you can judge it. CISA's table makes it high when Mission Prevalence is Essential — the vulnerable component "directly provides capabilities that constitute at least one MEF for at least one entity" (MEF: mission essential function) — or when Public Well-Being Impact is Irreversible: "multiple fatalities are likely", the cyber-physical system "is likely lost or destroyed", "extreme or serious externalities" are imposed on other parties, or social systems such as elections or the financial grid "are destabilized and potentially collapse". CISA's decision table
BOD 26-04 (CISA's remediation timeline, Table 1)60 days if the affected system is publicly exposed; Fix on system upgrade if it is not.
In CISA KEV no (CISA's KEV catalog) · Automatable no (CISA Vulnrichment) · Technical impact partial (CISA Vulnrichment). Whether your system is publicly exposed is yours to answer. BOD 26-04 binds US Federal Civilian Executive Branch agencies; for anyone else it is CISA's published timeline, a reference rather than an obligation. The directive · how EchelonGraph applies it
Published
August 17, 2026
Last Modified
October 2, 2026
Advisory Details (4)
Auto-updated Aug 17, 2026Derender endpoints decode caller-supplied GenerateResponse token IDs without output bounds · Advisory · vllm-project/vllm · GitHub
https://github.com/vllm-project/vllm/security/advisories/GHSA-8737-qx52-hjffv0.26.0
Patch available: vllm-project/vllm v0.26.0
https://github.com/vllm-project/vllm/releases/tag/v0.26.0fix(security): add resource bounds validation to derender endpoints
Fix merged in vllm-project/vllm PR #47260 on 2026-07-07 — awaiting tagged release
https://github.com/vllm-project/vllm/pull/47260commit 8e61b646e2d1 (vllm-project/vllm)
Fix landed in vllm-project/vllm commit 8e61b646e2d1 — awaiting tagged release
https://github.com/vllm-project/vllm/commit/8e61b646e2d157f9b93451fa048f9c8530c8a67bVendor Advisories for CVE-2026-71486(6)
These vendors published their own advisory mentioning this CVE — often with vendor-specific remediation steps + affected product lists not in NVD.
- RHSA-2026:82415Red Hat Product SecurityCritical
Red Hat Security Advisory: Red Hat AI Inference 3.5.2 (cpu)
- RHSA-2026:82371Red Hat Product SecurityCritical
Red Hat Security Advisory: Red Hat AI Inference 3.5.2 (spyre)
- RHSA-2026:82368Red Hat Product SecurityCritical
Red Hat Security Advisory: Red Hat AI Inference 3.5.2 (cuda)
- RHSA-2026:82061Red Hat Product SecurityCritical
Red Hat Security Advisory: Red Hat AI Inference Tech Preview 3.5.2 (Gaudi)
- RHSA-2026:82059Red Hat Product SecurityCritical
Red Hat Security Advisory: Red Hat AI Inference 3.5.2 (ROCm)
- GHSA-8737-qx52-hjffGitHub Security AdvisoriesMedium
vLLM: Derender endpoints decode caller-supplied GenerateResponse token IDs without output bounds
Affected Packages
(1 across 1 ecosystem)
PyPI(1)
| Package | Vulnerable range | Fix by version range | Dependents |
|---|---|---|---|
| vllm | 0.0.1 ... 0.9.2 (92 versions) |
| — |
Weakness Classification(2)
MITRE Common Weakness Enumeration — the root-cause categories this CVE belongs to.
Data Freshness Timeline
(refreshed 11× in last 7d / 45× in last 30d)
Each row is a time one of our pipelines fetched, updated or re-scored this CVE. For example, "NVD update" means NVD published or revised its analysis for this CVE; "MITRE cvelistV5" means we ingested or refreshed it from the CNA feed. An arrow marks a change the record states ("CVSS v3 → 9.1"); EPSS and GHSA rows show the values recorded at that time, not what changed (EPSS values are rounded; hover one for the exact figures). Most recent first.
A repeat, a row identical to one the same pipeline wrote less than 10 seconds earlier, is not a separate refresh and is not counted. 1 repeat row among the latest 93 is merged into the row it repeats.
- 2026-10-13 16:35 UTCEPSS rescoreEPSS 0.4% · 29th percentile
- 2026-10-12 16:34 UTCEPSS rescoreEPSS 0.4% · 29th percentile
- 2026-10-12 15:42 UTCEG score recompute
- 2026-10-11 19:28 UTCEG score recompute
- 2026-10-11 16:43 UTCEG score recompute
- 2026-10-11 14:03 UTCEG score recompute
- 2026-10-11 13:49 UTCEPSS rescoreEPSS 0.4% · 29th percentile
- 2026-10-10 20:00 UTCEPSS rescoreEPSS 0.4% · 29th percentile
- 2026-10-09 20:07 UTCEPSS rescoreEPSS 0.4% · 29th percentile
- 2026-10-08 17:13 UTCEPSS rescoreEPSS 0.4% · 29th percentile
- 2026-10-07 23:48 UTCEPSS rescoreEPSS 0.4% · 29th percentile
- 2026-10-07 11:26 UTCEG score recompute
- 2026-10-06 15:18 UTCEPSS rescoreEPSS 0.4% · 29th percentile
- 2026-10-06 09:02 UTCEG score recompute
- 2026-10-05 17:43 UTCEPSS rescoreEPSS 0.4% · 29th percentile
- 2026-10-04 23:22 UTCEPSS rescoreEPSS 0.4% · 28th percentile
- 2026-10-03 20:26 UTCEG score recompute
- 2026-10-03 14:26 UTCEPSS rescoreEPSS 0.4% · 29th percentile
- 2026-10-02 19:32 UTCEG score recompute
- 2026-10-02 15:45 UTCEG score recompute
- 2026-10-01 19:50 UTCEPSS rescoreEPSS 0.5% · 38th percentile
- 2026-10-01 10:53 UTCEG score recompute
- 2026-09-30 15:03 UTCEPSS rescoreEPSS 0.5% · 37th percentile
- 2026-09-30 09:58 UTCEG score recompute
- 2026-09-28 13:52 UTCEPSS rescoreEPSS 0.5% · 37th percentile
Show 67 moreShow fewer
- 2026-09-28 09:00 UTCEG score recompute
- 2026-09-27 13:48 UTCEPSS rescoreEPSS 0.5% · 37th percentile
- 2026-09-27 08:05 UTCEG score recompute
- 2026-09-26 15:59 UTCEPSS rescoreEPSS 0.5% · 37th percentile
- 2026-09-24 13:22 UTCEG score recompute
- 2026-09-23 17:54 UTCEPSS rescoreEPSS 0.3% · 27th percentile
- 2026-09-23 12:27 UTCEG score recompute
- 2026-09-22 11:33 UTCEG score recompute
- 2026-09-21 21:09 UTCEPSS rescoreEPSS 0.3% · 27th percentile
- 2026-09-21 10:38 UTCEG score recompute
- 2026-09-20 20:16 UTCEPSS rescoreEPSS 0.3% · 27th percentile
- 2026-09-20 09:44 UTCEG score recompute
- 2026-09-19 06:14 UTCEG score recompute
- 2026-09-18 19:28 UTCEPSS rescoreEPSS 0.3% · 27th percentile
- 2026-09-18 05:19 UTCEG score recompute
- 2026-09-17 04:25 UTCEG score recompute
- 2026-09-16 14:08 UTCEPSS rescoreEPSS 0.3% · 27th percentile
- 2026-09-16 05:15 UTCEPSS rescoreEPSS 0.3% · 27th percentile
- 2026-09-16 03:29 UTCEG score recompute
- 2026-09-15 03:12 UTCEPSS rescoreEPSS 0.3% · 27th percentile
- 2026-09-14 01:35 UTCEG score recompute
- 2026-09-13 00:40 UTCEG score recompute
- 2026-09-12 15:02 UTCEPSS rescoreEPSS 0.3% · 27th percentile
- 2026-09-11 23:45 UTCEG score recompute
- 2026-09-11 14:53 UTCEPSS rescoreEPSS 0.3% · 27th percentile
- 2026-09-10 22:22 UTCEG score recompute
- 2026-09-10 09:35 UTCEPSS rescoreEPSS 0.3% · 27th percentile
- 2026-09-09 03:38 UTCEG score recompute
- 2026-09-08 22:01 UTCEPSS rescoreEPSS 0.3% · 27th percentile
- 2026-09-07 16:11 UTCEG score recompute
- 2026-09-06 15:15 UTCEG score recompute
- 2026-09-06 13:48 UTCEPSS rescoreEPSS 0.3% · 27th percentile
- 2026-09-05 14:13 UTCEG score recompute
- 2026-09-04 13:10 UTCEG score recompute
- 2026-09-04 05:07 UTCEPSS rescoreEPSS 0.3% · 26th percentile
- 2026-09-03 12:14 UTCEG score recompute
- 2026-09-02 14:12 UTCEPSS rescoreEPSS 0.3% · 26th percentile
- 2026-09-02 10:38 UTCEG score recompute
- 2026-09-01 13:54 UTCEPSS rescoreEPSS 0.3% · 26th percentile
- 2026-09-01 09:44 UTCEG score recompute
- 2026-09-01 04:40 UTCEPSS rescoreEPSS 0.3% · 26th percentile
- 2026-08-31 07:25 UTCEG score recompute
- 2026-08-30 19:17 UTCEPSS rescoreEPSS 0.3% · 26th percentile
- 2026-08-30 06:20 UTCEG score recompute
- 2026-08-30 01:22 UTCEPSS rescoreEPSS 0.3% · 26th percentile
- 2026-08-29 05:24 UTCEG score recompute
- 2026-08-28 21:42 UTCEPSS rescoreEPSS 0.3% · 26th percentile
- 2026-08-27 21:43 UTCEG score recompute
- 2026-08-26 20:49 UTCEG score recompute
- 2026-08-26 14:47 UTCEPSS rescoreEPSS 0.3% · 26th percentile
- 2026-08-25 19:53 UTCEG score recompute
- 2026-08-25 13:49 UTCEPSS rescoreEPSS 0.3% · 26th percentile
- 2026-08-24 18:58 UTCEG score recompute
- 2026-08-23 18:04 UTCEG score recompute
- 2026-08-23 00:19 UTCEPSS rescoreEPSS 0.3% · 27th percentile
- 2026-08-22 17:09 UTCEG score recompute
- 2026-08-21 23:49 UTCEPSS rescoreEPSS 0.3% · 27th percentile
- 2026-08-21 16:11 UTCEG score recompute
- 2026-08-20 22:56 UTCEPSS rescoreEPSS 0.3% · 27th percentile
- 2026-08-20 15:13 UTCEG score recompute
- 2026-08-19 17:04 UTCEPSS rescoreEPSS 0.3% · 27th percentile
- 2026-08-19 14:19 UTCEG score recompute
- 2026-08-18 13:49 UTCEPSS rescoreEPSS 0.3% · 27th percentile
- 2026-08-18 13:23 UTCEG score recompute
- 2026-08-18 12:42 UTCEG score recompute
- 2026-08-17 20:20 UTCEG score recomputeFirst assessed: 4.3 MEDIUM
- 2026-08-17 20:19 UTCMITRE cvelistV5first tracked
Related CVEs(same product + same CWE)
Same product
10 shownPyPI:vllm
Frequently asked(5)
What is CVE-2026-71486?
When was the CVE record for CVE-2026-71486 published?
Is CVE-2026-71486 actively exploited?
What is the CVSS score of CVE-2026-71486?
How do I remediate CVE-2026-71486?
Dependency Blast Radius
See which npm, PyPI, Go, and Maven packages are affected by CVE-2026-71486
Is Your Infrastructure Affected by CVE-2026-71486?
EchelonGraph automatically scans your cloud infrastructure and maps CVE exposure using blast radius analysis.