PyPI / vllm

vLLM allows Remote Code Execution by Pickle Deserialization via AsyncEngineRPCServer() RPC server entrypoints

수정: 2026. 6. 8.

CVE-2025-24357 Malicious model remote code execution fix bypass with PyTorch < 2.6.0

수정: 2026. 2. 4.

vLLM is vulnerable to DoS in Idefics3 vision models via image payload with ambiguous dimensions

수정: 2026. 6. 8.

vLLM vulnerable to Denial of Service by abusing xgrammar cache

수정: 2025. 4. 15.

vLLM Vulnerable to Remote Code Execution via Mooncake Integration

수정: 2026. 2. 4.

vLLM Allows Remote Code Execution via PyNcclPipe Communication Service

수정: 2026. 2. 4.

vLLM Vulnerable to Remote DoS via Special-Token Placeholders

수정: 2026. 6. 1.

vLLM vulnerable to Regular Expression Denial of Service

수정: 2026. 2. 4.

vLLM introduced enhanced protection for CVE-2025-62164

수정: 2026. 2. 3.

vLLM denial of service via outlines unbounded cache on disk

수정: 2026. 6. 8.

vLLM deserialization vulnerability leading to DoS and potential RCE

수정: 2026. 2. 4.

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

수정: 2026. 4. 10.

vLLM deserialization vulnerability in vllm.distributed.GroupCoordinator.recv_object

수정: 2026. 2. 4.

vLLM vulnerable to DoS with incorrect shape of multimodal embedding inputs

수정: 2026. 2. 4.

vLLM: Denial of Service via Unbounded Frame Count in video/jpeg Base64 Processing

수정: 2026. 6. 8.

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

수정: 2026. 3. 27.

vllm: Malicious model to RCE by torch.load in hf_model_weights_iterator

수정: 2026. 2. 4.

vLLM uses Python 3.12 built-in hash() which leads to predictable hash collisions in prefix cache

수정: 2026. 2. 4.

vllm API endpoints vulnerable to Denial of Service Attacks

수정: 2026. 2. 4.

vLLM has SSRF Protection Bypass

수정: 2026. 3. 16.

phi4mm: Quadratic Time Complexity in Input Token Processing​ leads to denial of service

수정: 2026. 2. 4.

vLLM Tool Schema allows DoS via Malformed pattern and type Fields

수정: 2026. 2. 4.

vLLM denial of service vulnerability

수정: 2026. 2. 4.

vLLM has a Regular Expression Denial of Service (ReDoS, Exponential Complexity) Vulnerability in `pythonic_tool_parser.py`

수정: 2026. 2. 4.

vLLM Denial of Service via the best_of parameter

수정: 2026. 2. 4.

vLLM is vulnerable to timing attack at bearer auth

수정: 2026. 2. 4.

vLLM makes Use of Uninitialized Resource

수정: 2026. 6. 1.

vLLM Allows Remote Code Execution via Mooncake Integration

수정: 2026. 2. 4.

vLLM affected by RCE via auto_map dynamic module loading during model initialization

수정: 2026. 2. 4.

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

수정: 2026. 2. 4.

vLLM: Unauthenticated OOM Denial of Service via Unbounded `n` Parameter in OpenAI API Server

수정: 2026. 4. 10.

Potential Timing Side-Channel Vulnerability in vLLM’s Chunk-Based Prefix Caching

수정: 2026. 2. 4.

vLLM has RCE In Video Processing

수정: 2026. 3. 27.

vLLM Deserialization of Untrusted Data vulnerability

수정: 2026. 2. 4.

vLLM vulnerable to DoS via large Chat Completion or Tokenization requests with specially crafted `chat_template_kwargs`

수정: 2026. 2. 4.

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

수정: 2026. 2. 4.

vLLM DOS: Remotely kill vllm over http with invalid JSON schema

수정: 2026. 2. 4.

vLLM has Hardcoded Trust Override in Model Files Enables RCE Despite Explicit User Opt-Out

수정: 2026. 3. 30.

vLLM has remote code execution vulnerability in the tool call parser for Qwen3-Coder

수정: 2026. 2. 4.

vLLM: extract_hidden_states speculative decoding crashes server on any request with penalty parameters

수정: 2026. 6. 8.

vLLM vulnerable to remote code execution via transformers_utils/get_config

수정: 2025. 12. 2.

Data exposure via ZeroMQ on multi-node vLLM deployment

수정: 2026. 2. 4.

vLLM allows clients to crash the openai server with invalid regex

수정: 2026. 2. 4.

Remote Code Execution Vulnerability in vLLM Multi-Node Cluster Configuration

수정: 2026. 2. 4.

vLLM has a Weakness in MultiModalHasher Image Hashing Implementation

수정: 2026. 2. 4.

수정: 2026. 5. 21.

수정: 2026. 5. 20.

수정: 2026. 2. 4.

수정: 2025. 5. 29.

수정: 2026. 2. 5.

수정: 2025. 6. 26.

수정: 2025. 6. 26.

수정: 2025. 6. 26.

수정: 2025. 6. 27.

수정: 2025. 7. 1.

수정: 2025. 7. 1.

수정: 2026. 5. 20.

수정: 2026. 5. 20.

수정: 2026. 5. 20.