CVE-2026-41523High· 7.5▾ TwilightA flaw was found in vLLM, an inference and serving engine for large language models (LLMs). An unauthenticated attacker can exploit an assert-based security check during activation function loading. By publishing a malicious HuggingFace mo…
▾ Twilight zone — High severity, or a signal on a lesser flaw
impact 41.3 · likelihood 0.2 · exploitation 0
Need a working PoC? Pro members can cast a request and our team develops one — it lands right here.
Exploit-prediction probability, daily snapshots since Aug 2.
Disclosure to exploitation, from the record and what we observed since indexing it.
Disclosed via CSAF
0.7%
0.7% → 0.9%
Last analysed / modified upstream
A flaw was found in vLLM, an inference and serving engine for large language models (LLMs). An unauthenticated attacker can exploit an assert-based security check during activation function loading. By publishing a malicious HuggingFace model, an attacker can achieve arbitrary code execution on the server when vLLM runs in Python optimized mode.
vllm: vLLM: Arbitrary code execution via malicious HuggingFace model — rated Important by Red Hat. Released 2026-06-22, updated 2026-10-07.
Affected:
Fixed:
No fix planned:
Not affected:
For more information visit https://access.redhat.com/errata/RHSA-2026:61627 https://access.redhat.com/errata/RHSA-2026:61627 For more information visit https://access.redhat.com/errata/RHSA-2026:36005 https://access.redhat.com/errata/RHSA-2026:36005 For more information visit https://access.redhat.com/errata/RHSA-2026:61629 https://access.redhat.com/errata/RHSA-2026:61629
Workarounds / mitigations:
Affected packages:
vllm < 0.22.0Patched in:
vllm 0.22.0Connected by shared product, vendor, weakness, or advisory.
CVE-2026-45804High· 7.5diffusers: Diffusers: Arbitrary code execution due to trust_remote_code guard bypass (CVE-2026-45804)
CVE-2026-54234High· 7.5vllm: vLLM: Denial of Service via malformed speculative decoding workload (CVE-2026-54234)
CVE-2026-55574High· 7.5vllm: vLLM: Denial of Service via adversarial regular expression in structured outputs API (CVE-2026-55574)
CVE-2026-105754Medium· 6.5vLLM is an inference and serving engine for large language models
CVE-2026-105753Medium· 6.5vLLM is an inference and serving engine for large language models
CVE-2025-59530Medium· 5.3github.com/quic-go/quic-go: quic-go Crash Due to Premature HANDSHAKE_DONE Frame (CVE-2025-59530)