HomeCVE Intelligence › CVE-2026-41523
CVSS 7.5 HIGH Vulnerability

CVE-2026-41523: vLLM: Security Check Bypass via assert Statement in Activation Function Loading Allows Ar…

Summary 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 r…

7.5CVSS Score
HIGHSeverity
NOCISA KEV
VulnerabilityImpact Type

📋 Vulnerability Details

CVE IDCVE-2026-41523
Vendorpip
Affected Productvllm
Vulnerability TypeVulnerability
CVSS Score7.5 (HIGH)
Actively Exploited❌ No known exploitation
Patch StatusSee Vendor Advisory →
Reported ByCYBERDUDEBIVASH SENTINEL APEX Intelligence (via github_advisories)

🔬 Technical Analysis

#

Summary 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).

Details vLLM uses an assert statement at [vllm/model_executor/layers/pooler/activations.py:48](https://github.com/vllm-project/vllm/blob/main/vllm/model_executor/layers/pooler/activations.py#L48) as its sole security control to restrict which activation functions can be loaded from a HuggingFace model's config.json: ```python

vllm/model_executor/layers/pooler/activations.py:35-53

function_name: str | None = None if ( hasattr(config, "sentence_transformers") and "activation_fn" in config.sente

🎯 Known Indicators of Compromise

{"type":"url","value":"https://github.com/vllm-project/vllm/blob/main/vllm/model_executor/layers/pooler/activations.py#L48)","confidence_score":0.82,"first_seen":"2026-06-16","source_count":1}

📚 Advisory References

⚡ DETECTION RULES AVAILABLE

Get CVE-2026-41523 Detection Pack

Sigma rules, YARA signatures, IOC table, and SIEM queries for Splunk, Elastic, Sentinel, and Chronicle — deployable in 5 minutes.

✓ Sigma Rules ✓ YARA Pack ✓ IOC Table ✓ SIEM Queries
🛡️ Get Detection Pack → 🔌 Access via API →

🔗 Related Intelligence