PYSEC-2026-2300

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 vLL...

CVE-2026-47155

vLLM is an inference and serving engine for large language models LLMs. Prior to 0.22.0, vLLM's revision pinning controls do not consistently apply to all artifacts loaded for a model. A deployment that supplies --revision or --code-revision can still load dynamic code, GGUF files, image...

CVE-2026-54233

Affected software: vLLM (inference/serving engine). Vulnerability: decoding an audio file on the /v1/audio/transcriptions endpoint can cause extreme memory growth. A 25 MB OPUS upload decodes to about 14.9 GB of float32 PCM, because the audio decoder concatenates all frames in memory before retur...

CVE-2026-54235 vLLM: temperature=NaN and temperature=Infinity bypass validation and propagate to GPU kernels

vLLM is an inference and serving engine for large language models LLMs. Prior to 0.23.1rc0, ll temperature validation gates use comparison operators , which silently evaluate to False for NaN and for positive Infinity in Python's IEEE 754 float semantics. Both values pass every guard and propagat...

EUVD-2026-38400

vLLM is an inference and serving engine for large language models LLMs. From 0.5.5 until 0.23.1rc0, integer truncation of tensor dimensions in vLLM's GGUF dequantize kernels csrc/quantization/gguf/ggufkernel.cu causes partial tensor processing. The output tensor is allocated at full size via...

CVE-2026-53923

Summary of CVE-2026-53923 : The vulnerability affects vLLM (GGUF dequantize kernels) where integer truncation of tensor dimensions causes partially filled output tensors. From 0.5.5 up to 0.23.1rc0, the code allocates the full output tensor (torch::empty) but the CUDA kernel processes only a trun...

CVE-2026-53923 vLLM GGUF Kernels: int64_t to int truncation of tensor dimensions causes GPU buffer overflow

vLLM is an inference and serving engine for large language models LLMs. From 0.5.5 until 0.23.1rc0, integer truncation of tensor dimensions in vLLM's GGUF dequantize kernels csrc/quantization/gguf/ggufkernel.cu causes partial tensor processing. The output tensor is allocated at full size via...

crypto/x509: golang: Denial of Service due to excessive resource consumption via crafted certificate

A flaw was found in golang. A remote attacker could exploit this vulnerability by providing a specially crafted certificate during the error string construction process within the HostnameError.Error function. This flaw, caused by unbounded string concatenation, leads to excessive resource...

crypto/x509: golang: Go crypto/x509: Certificate validation bypass due to incorrect DNS constraint application

A flaw was found in the crypto/x509 package within Go golang. When verifying a certificate chain, excluded DNS Domain Name System constraints are not correctly applied to wildcard DNS Subject Alternative Names SANs if the case of the SAN differs from the constraint. This oversight could allow an...

crypto/x509: crypto/tls: golang: Go: Denial of Service vulnerability in certificate chain building

A flaw was found in the Go standard library packages crypto/x509 and crypto/tls. During the process of building a certificate chain, an attacker can provide a large number of intermediate certificates. This excessive input is not properly limited, leading to an uncontrolled amount of work being...

crypto/tls: golang: Go crypto/tls: Denial of Service via multiple TLS 1.3 key update messages

A flaw was found in the crypto/tls package within the Go golang standard library, specifically affecting TLS 1.3 connections. A remote attacker can exploit this vulnerability by sending multiple key update messages in a single record after the handshake. This can cause the connection to deadlock,...

golang: internal/syscall/unix: Root.Chmod can follow symlinks out of the root

A flaw was found in the internal/syscall/unix package in the Go standard library. If the target of the Root.Chmod function is replaced with a symbolic link during execution, specifically after Root.Chmod checks the target but before acting, the chmod operation will be performed on the file the...

CVE-2026-49468 LiteLLM: Authentication Bypass via Host Header Injection

LiteLLM is a proxy server AI Gateway to call LLM APIs in OpenAI or native format. Prior to 1.84.0, a Host-header parsing flaw in the LiteLLM proxy could, under specific conditions, allow unauthenticated access to protected management routes. The auth layer derived the effective route from...

CVE-2026-54293

NLTK Natural Language Toolkit is a suite of open source Python modules, data sets, and tutorials supporting research and development in Natural Language Processing. Prior to 3.10.0-rc1, nltk.data.load in NLTK is vulnerable to path traversal via URL-encoded path separators and traversal segments...

PYSEC-2026-2078

NLTK Natural Language Toolkit is a suite of open source Python modules, data sets, and tutorials supporting research and development in Natural Language Processing. Prior to 3.10.0-rc1, nltk.data.load in NLTK is vulnerable to path traversal via URL-encoded path separators and traversal segments...

UBUNTU-CVE-2026-54293

NLTK Natural Language Toolkit is a suite of open source Python modules, data sets, and tutorials supporting research and development in Natural Language Processing. Prior to 3.10.0-rc1, nltk.data.load in NLTK is vulnerable to path traversal via URL-encoded path separators and traversal segments...

CVE-2026-44271

Dell Wyse Management Suite WMS, versions prior to WMS 2605, contain an Improper Neutralization of Special Elements used in an SQL Command 'SQL Injection' vulnerability. A low privileged attacker with remote access could potentially exploit this vulnerability, leading to Unauthorized access...

CVE-2026-53779 WebP Server Go < 0.15.0 Path Traversal via Backslash Encoding on Windows

WebP Server Go through 0.14.4 contains a path traversal vulnerability on Windows that allows unauthenticated attackers to read files outside the configured IMGPATH directory by sending requests with percent-encoded backslashes %5C that bypass the path.Clean sanitization in handler/router.go...

CVE-2026-54293

CVE-2026-54293 affects NLTK’s nltk.data.load() in Python. A TOCTOU-style flaw lets an attacker bypass the unsafe-path regex (UNSAFE_NO_PROTOCOL_RE) by using URL-encoded path separators (e.g., %2f, %2e%2e) and then decoding, enabling arbitrary local file reads prior to the fix. Affected until vers...

EUVD-2026-38333

NLTK Natural Language Toolkit is a suite of open source Python modules, data sets, and tutorials supporting research and development in Natural Language Processing. Prior to 3.10.0-rc1, nltk.data.load in NLTK is vulnerable to path traversal via URL-encoded path separators and traversal segments...