PYSEC-2025-62

vLLM is a high-throughput and memory-efficient inference and serving engine for LLMs. Maliciously constructed statements can lead to hash collisions, resulting in cache reuse, which can interfere with subsequent responses and cause unintended behavior. Prefix caching makes use of Python's built-i...

CVE-2025-25183

CVE-2025-25183 affects vLLM (prefix cache) where malicious inputs can trigger Python 3.12’s hash(None) behaving as a predictable constant, enabling hash collisions in the prefix cache. This may allow cache entries created from one prompt to be reused for another, causing unintended behavior in re...

CVE-2025-25183 vLLM using built-in hash() from Python 3.12 leads to predictable hash collisions in vLLM prefix cache

vLLM is a high-throughput and memory-efficient inference and serving engine for LLMs. Maliciously constructed statements can lead to hash collisions, resulting in cache reuse, which can interfere with subsequent responses and cause unintended behavior. Prefix caching makes use of Python's built-i...

vLLM 安全漏洞

vLLM is a high throughput and memory efficient inference and service engine for LLM from the vLLM open source. A security vulnerability exists in vLLM that stems from a maliciously constructed statement that could lead to a hash collision, which could lead to cache reuse, which could interfere wi...

GHSA-RM76-4MRF-V9R8 vLLM uses Python 3.12 built-in hash() which leads to predictable hash collisions in prefix cache

Summary Maliciously constructed prompts can lead to hash collisions, resulting in prefix cache reuse, which can interfere with subsequent responses and cause unintended behavior. Details vLLM's prefix caching makes use of Python's built-in hash function. As of Python 3.12, the behavior of hashNon...

PT-2025-6000 · Vllm +1 · Vllm +1

Name of the Vulnerable Software and Affected Versions: vLLM versions prior to 0.7.2 Description: Maliciously constructed statements can lead to hash collisions, resulting in cache reuse, which can interfere with subsequent responses and cause unintended behavior. The issue arises from the use of...

Design/Logic Flaw

When loading a script with Subresource Integrity, attackers with an injection capability could trigger the reuse of previously cached entries with incorrect, different integrity metadata. This vulnerability affects Firefox 103...

CVE-2022-36315

When loading a script with Subresource Integrity, attackers with an injection capability could trigger the reuse of previously cached entries with incorrect, different integrity metadata. This vulnerability affects Firefox 103...

CVE-2022-36315

When loading a script with Subresource Integrity, attackers with an injection capability could trigger the reuse of previously cached entries with incorrect, different integrity metadata. This vulnerability affects Firefox 103...

The vulnerability of the Apache Tomcat application server allows attackers to obtain confidential information.

The vulnerability of the Apache Tomcat application server is related to improper data processing. Exploiting this vulnerability allows a malicious actor to obtain confidential information from a remote perspective. If the processing of sent files is completed quickly, it is possible that the same...

USN-3275-1 openjdk-8 vulnerabilities

It was discovered that OpenJDK improperly re-used cached NTLM connections in some situations. A remote attacker could possibly use this to cause a Java application to perform actions with the credentials of a different user. CVE-2017-3509 It was discovered that an untrusted library search path fl...

BEA产品多个远程安全漏洞

BEA Systems WebLogic包含多种应用系统集成方案，包括Server/Express/Integration等。 BEA Weblogic中存在多个安全漏洞，可能允许恶意攻击者获得敏感信息、绕过某些安全限制、导致拒绝服务或完全入侵系统。 这些漏洞包括： 1 SSL库中的漏洞可能允许判断明文块； 2 从缓存重用连接时服务器没有正确地验证客户端证书，导致攻击者可以通过X.509证书访问Web服务器。成功攻击要求应用程序允许通过单个客户端进程访问多个用户； 3 存储在JDBCDataSourceFactory MBean Properties属性中的口令没有加密； 4...