vLLM <= 0.23.0 - Anthropic Router Heap Address Information Leak

vLLM = 0.23.0 incompletely fixes CVE-2026-22778. The original fix added sanitizemessage to the OpenAI router but the Anthropic-compatible router /v1/messages echoes strexc directly. id: CVE-2026-54236 info: name: vLLM = 0.23.0 - Anthropic Router Heap Address Information Leak author: kenlacroix...

PYSEC-2026-3542 vLLM has Remote DoS via Invalid Recovered Token Reinjection

Summary A frontend-legal multi-request speculative workload can make vLLM produce an out-of-vocabulary recovered token equal to vocabsize, convert that value to -1 when choosing the next live token for a request, and then feed that -1 back into the next drafter input ids. On Qwen3 GPTQ this reach...

GHSA-33CG-GXV8-3P8G vLLM denial of service via prompt embeds on M-RoPE models

Summary Short summary of the problem. Make the impact and severity as clear as possible. For example: An unsafe deserialization vulnerability allows any unauthenticated user to execute arbitrary code on the server. Sending a pure prompt embeds payload in a /v1/completions request with a model usi...

GHSA-V82G-2437-67M2 vLLM: Speech-to-text upload size limit is enforced after full UploadFile read

Summary Current-head vLLM documents VLLMMAXAUDIOCLIPFILESIZEMB as the maximum audio file size accepted by the speech-to-text APIs. The default is 25 MB. vllm/envs.py also describes files larger than this value as rejected. The /v1/audio/transcriptions and /v1/audio/translations routes call await...

PYSEC-2026-3405 vLLM: temperature=NaN and temperature=Infinity bypass validation and propagate to GPU kernels

Summary All 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 propagate to GPU sampling kernels, where they produce undefined behavior or CUDA errors tha...

PYSEC-2026-3403 vLLM: GGUF dequantize kernel int truncation exposes uninitialized GPU memory in multi-tenant serving

Summary 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 torch::empty uninitialized memory, but the dequantize CUDA kernel processes only a truncated...

PYSEC-2026-3409 vLLM Vulnerable to Remote DoS via Special-Token Placeholders

Summary This report explains a Token Injection vulnerability in vLLM’s multimodal processing. Unauthenticated, text-only prompts that spell special tokens are interpreted as control. Image and video placeholder sequences supplied without matching data cause vLLM to index into empty grids during...

PYSEC-2026-3410 vLLM: Server-Side Request Forgery (SSRF) in `download_bytes_from_url `

Summary A Server Side Request Forgery SSRF vulnerability in downloadbytesfromurl allows any actor who can control batch input JSON to make the vLLM batch runner issue arbitrary HTTP/HTTPS requests from the server, without any URL validation or domain restrictions. This can be used to target...

PYSEC-2026-3411 vLLM has SSRF Protection Bypass

Summary The SSRF protection fix for https://github.com/vllm-project/vllm/security/advisories/GHSA-qh4c-xf7m-gxfc can be bypassed in the loadfromurlasync method due to inconsistent URL parsing behavior between the validation layer and the actual HTTP client. Affected Component - File:...

EUVD-2026-43310

A flaw was found in the vllm-orchestrator-gateway component. The system's production binary logs all incoming authorization headers and full chat payloads, which may contain personally identifiable information PII and secrets, to persistent logs. This sensitive data, including bearer tokens and...

CVE-2026-15574

A flaw was found in the vllm-orchestrator-gateway component. The system's production binary logs all incoming authorization headers and full chat payloads, which may contain personally identifiable information PII and secrets, to persistent logs. This sensitive data, including bearer tokens and...

CVE-2026-15574

The CVE-2026-15574 issue affects the vllm-orchestrator-gateway component, where the production binary logs incoming authorization headers and full chat payloads, potentially exposing PII, secrets, bearer tokens, and conversation content. This data exposure arises from a hard-coded debug/default l...

CVE-2026-15574

A flaw was found in the vllm-orchestrator-gateway component. The system's production binary logs all incoming authorization headers and full chat payloads, which may contain personally identifiable information PII and secrets, to persistent logs. This sensitive data, including bearer tokens and...

CVE-2026-55646

A flaw was found in vLLM, an inference and serving engine for large language models. A remote attacker, by submitting an oversized audio file to the /v1/audio/transcriptions or /v1/audio/translations routes, can cause the system to allocate excessive memory. This occurs before the configured file...

CVE-2026-54234

A flaw was found in vLLM, a high-throughput and memory-efficient inference and serving engine for Large Language Models LLMs. A remote attacker can exploit this vulnerability by sending a specially crafted multi-request speculative decoding workload through public gRPC Generate and Abort endpoint...

PYSEC-2026-2020 vLLM vulnerable to Server-Side Request Forgery (SSRF) through MediaConnector

Summary A Server-Side Request Forgery SSRF vulnerability exists in the MediaConnector class within the vLLM project's multimodal feature set. The loadfromurl and loadfromurlasync methods obtain and process media from URLs provided by users, using different Python parsing libraries when restrictin...

PYSEC-2026-2018 vLLM deserialization vulnerability leading to DoS and potential RCE

Summary A memory corruption vulnerability that leading to a crash denial-of-service and potentially remote code execution RCE exists in vLLM versions 0.10.2 and later, in the Completions API endpoint. When processing user-supplied prompt embeddings, the endpoint loads serialized tensors using...

PYSEC-2026-2011 vLLM is vulnerable to Server-Side Request Forgery (SSRF) through `MediaConnector` class

Summary A Server-Side Request Forgery SSRF vulnerability exists in the MediaConnector class within the vLLM project's multimodal feature set. The loadfromurl and loadfromurlasync methods fetch and process media from user-provided URLs without adequate restrictions on the target hosts. This allows...

PYSEC-2026-2026 vLLM is vulnerable to timing attack at bearer auth

Summary The API key support in vLLM performed validation using a method that was vulnerable to a timing attack. This could potentially allow an attacker to discover a valid API key using an approach more efficient than brute force. Details...

PYSEC-2026-2013 vLLM: Resource-Exhaustion (DoS) through Malicious Jinja Template in OpenAI-Compatible Server

Summary A resource-exhaustion denial-of-service vulnerability exists in multiple endpoints of the OpenAI-Compatible Server due to the ability to specify Jinja templates via the chattemplate and chattemplatekwargs parameters. If an attacker can supply these parameters to the API, they can cause a...