vLLM is an inference and serving engine for large language models (LLMs). Prior to 0.22.0, an assert-based security check in vLLM's activation function loading allows any unauthenticated attacker to achieve arbitrary code execution on the server by publishing a malicious HuggingFace model, when vLLM runs in Python optimized mode (python -O or PYTHONOPTIMIZE=1). This vulnerability is fixed in 0.22.0.
CVE-2026-41523
This high-severity CVE scores 7.5 under the CNA's CVSS (NVD's own analysis pending). EPSS exploit probability: 0.5%, top 61% of all CVEs by exploit prediction. GitHub Security Advisory data not yet ingested — confidence will rise once GHSA publishes (typical lag: hours to days for open-source ecosystem CVEs; never for infrastructure-only CVEs).
- High severity, but no confirmed exploitation yet
A fix is available — apply it.
- CVSS v3
- 7.5
- EG Score
- 7.5(medium)
- EG Risk
- 34(Track)EG Risk 34/100SSVC: Track
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 computedSeverity75% × 45%Exploitation0% × 40%Automatability0% × 15%Action: Routine — remediate on your standard cadence. - EPSS
- 38.9%
- KEV
- Not listed
Published
June 16, 2026
Last Modified
July 22, 2026
Advisory Details (3)
Auto-updated Jun 23, 2026huntr - The world's first bug bounty platform for AI/ML
https://huntr.com/bounties/dcb05b04-e625-41e7-adbc-bbae0cc2d64ccommit b3c7ffcab82c (vllm-project/vllm)
Fix landed in vllm-project/vllm commit b3c7ffcab82c — awaiting tagged release
https://github.com/vllm-project/vllm/commit/b3c7ffcab82c2439726f8cb213800f6f38c023d3Security Check Bypass via assert Statement in Activation Function Loading Allows Arbitrary Code Execution · Advisory · vllm-project/vllm · GitHub
https://github.com/vllm-project/vllm/security/advisories/GHSA-q8gq-377p-jq3rVendor Advisories for CVE-2026-41523(3)
These vendors published their own advisory mentioning this CVE — often with vendor-specific remediation steps + affected product lists not in NVD.
- RHSA-2026:36006Red Hat Product SecurityHigh
Red Hat Security Advisory: Red Hat AI Inference Server 3.2.2 (ROCm)
- RHSA-2026:36005Red Hat Product SecurityHigh
Red Hat Security Advisory: Red Hat AI Inference Server 3.2.2 (CUDA)
- GHSA-q8gq-377p-jq3rGitHub Security AdvisoriesHigh
vLLM: Security Check Bypass via assert Statement in Activation Function Loading Allows Arbitrary Code Execution
Patch Availability(2)
| Vendor / Ecosystem | Fixed in / Patch | Released | Source |
|---|---|---|---|
| redhat | rhaiis/vllm-cuda-rhel9:1782951012 | 2026-07-06 | redhat |
| redhat | rhaiis/vllm-rocm-rhel9:1782951244 | 2026-07-06 | redhat |
Patches are aggregated from vendor advisories (Red Hat, Microsoft, Cisco, GitHub) and package ecosystems (OSV, GHSA). Multiple rows for the same upstream release have been deduplicated.
Affected Packages
(1 across 1 ecosystem)
PyPI(1)
| Package | Vulnerable range | Fixed in | Dependents |
|---|---|---|---|
| vllm | 0.0.1 ... 0.9.2 (86 versions) | 0.22.0 | — |
Weakness Classification(2)
MITRE Common Weakness Enumeration — the root-cause categories this CVE belongs to.
Data Freshness Timeline
(refreshed 9× in last 7d / 54× in last 30d)
Each row is a source pipeline that fetched or updated this CVE on that date, with what changed. 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. Most recent first.
- 2026-07-23 03:11 UTCEG score recompute
- 2026-07-22 14:08 UTCEPSS rescore
- 2026-07-21 15:24 UTCEPSS rescore
- 2026-07-20 17:08 UTCEPSS rescore
- 2026-07-19 14:31 UTCEPSS rescore
- 2026-07-19 02:29 UTCEPSS rescore
- 2026-07-18 10:04 UTCEPSS rescore
- 2026-07-16 22:40 UTCVendor advisory
- 2026-07-16 17:03 UTCEPSS rescore
- 2026-07-16 03:14 UTCVendor advisory
- 2026-07-15 16:57 UTCEPSS rescore
- 2026-07-15 16:57 UTCEPSS rescore
- 2026-07-15 15:28 UTCVendor advisory
- 2026-07-15 02:00 UTCEPSS rescore
- 2026-07-15 01:17 UTCVendor advisory
- 2026-07-14 11:44 UTCVendor advisory
- 2026-07-13 22:50 UTCVendor advisory
- 2026-07-13 22:30 UTCEPSS rescore
- 2026-07-13 11:05 UTCVendor advisory
- 2026-07-13 06:13 UTCEPSS rescore
- 2026-07-13 06:13 UTCEPSS rescore
- 2026-07-12 22:14 UTCVendor advisory
- 2026-07-12 10:29 UTCVendor advisory
- 2026-07-12 05:46 UTCEPSS rescore
- 2026-07-11 22:43 UTCVendor advisory
Show 30 moreShow fewer
- 2026-07-11 10:58 UTCVendor advisory
- 2026-07-11 08:27 UTCEPSS rescore
- 2026-07-10 23:11 UTCVendor advisory
- 2026-07-10 11:19 UTCVendor advisory
- 2026-07-09 22:52 UTCVendor advisory
- 2026-07-09 19:10 UTCEPSS rescore
- 2026-07-09 11:04 UTCVendor advisory
- 2026-07-08 23:04 UTCVendor advisory
- 2026-07-08 15:15 UTCEPSS rescore
- 2026-07-08 11:17 UTCVendor advisory
- 2026-07-07 23:29 UTCVendor advisory
- 2026-07-07 13:46 UTCEPSS rescore
- 2026-07-07 11:44 UTCVendor advisory
- 2026-07-06 16:27 UTCEPSS rescore
- 2026-07-06 16:27 UTCEPSS rescore
- 2026-07-06 02:23 UTCEPSS rescore
- 2026-07-06 02:23 UTCEPSS rescore
- 2026-07-05 02:30 UTCEPSS rescore
- 2026-07-04 06:31 UTCEPSS rescore
- 2026-07-01 15:06 UTCEPSS rescore
- 2026-06-30 23:22 UTCEPSS rescore
- 2026-06-29 14:06 UTCEPSS rescore
- 2026-06-28 14:07 UTCEPSS rescore
- 2026-06-28 04:56 UTCEPSS rescore
- 2026-06-27 03:08 UTCEPSS rescore
- 2026-06-25 13:49 UTCEPSS rescore
- 2026-06-25 13:49 UTCEPSS rescore
- 2026-06-24 14:05 UTCEPSS rescore
- 2026-06-23 21:33 UTCEPSS rescore
- 2026-06-16 17:47 UTCEG score recompute
Related CVEs(same vendor + same CWE)
Same vendor
10 shownredhat
Same CWE
10 shownCWE-94
Frequently asked(5)
What is CVE-2026-41523?
When was CVE-2026-41523 disclosed?
Is CVE-2026-41523 actively exploited?
What is the CVSS score of CVE-2026-41523?
How do I remediate CVE-2026-41523?
Dependency Blast Radius
See which npm, PyPI, Go, and Maven packages are affected by CVE-2026-41523
Is Your Infrastructure Affected by CVE-2026-41523?
EchelonGraph automatically scans your cloud infrastructure and maps CVE exposure using blast radius analysis.