CVE-2026-114049
LOWDescription
vLLM before 0.31.0 contains a cross-request mask leak in InputBatch.condense() that allows remote attackers to constrain other clients' outputs by leaving stale allowed_token_ids rows. Attackers can submit concurrent constrained generate requests, including via unauthenticated /invocations, so recycled batch slots apply their token allowlist to victims' unrestricted requests.
Is your site exposed to CVE-2026-114049?
Run a free security scan — no signup, results in seconds.
CVSS v3.1 Score
EPSS — Exploit Prediction
EPSS estimates the probability that this vulnerability will be exploited in the wild within the next 30 days. A higher score means more likely to be exploited.
Weakness Type (CWE)
References
Frequently Asked Questions
What is CVE-2026-114049? +
How severe is CVE-2026-114049? +
How do I check if I'm vulnerable to CVE-2026-114049? +
Related Vulnerabilities
Expected behavior violation in AMD Secure Encrypted Virtualization (SEV) firmware could allow a malicious hypervisor to bypass Reverse Map Table …
urllib3 is an HTTP client library for Python. From 1.26.0 until 2.8.0, the proxy_ssl_context, proxy_assert_hostname, proxy_assert_fingerprint, ssl_context, cert_reqs, verify_mode, use_forwarding_for_https=True, …
An unencrypted 'pause encryption request' message causes a denial of service in the in the RS9116W/SiWx917. See vulnerability B-E10 in …
On affected Arista EOS platforms with Routing Information Protocol version 2 (RIPv2) configured, under certain conditions, RIPv2 might be silently …
Apollo Router is a configurable, graph router written in Rust to run a federated supergraph that uses Apollo Federation 2. …
In danny-avila/librechat version 0.7.9, there is an insecure API design issue in the 2-Factor Authentication (2FA) flow. The system allows …