Data

Tools

Indexes

Applications

Experiments

Advisories

An open API service providing security vulnerability metadata for many open source software ecosystems.

pypi

vllm

pypi

A high-throughput and memory-efficient inference and serving engine for LLMs

View on github.com · View on pypi.org

Security Advisories for vllm in pypi

Moderate
6 days ago

vLLM: Request-selected PyNvVideoCodec GPU decode bypasses static VRAM reservation GSA_kwCzR0hTQS04cHcyLTZqdjMtbWo1as4AB5OW

pypi vllm
Moderate
6 days ago

vLLM: Unauthenticated audio decompression-bomb DoS in /v1/chat/completions GSA_kwCzR0hTQS1oY3dxLTh3amYtM2djcs4AB27I

pypi vllm
Moderate
14 days ago

vLLM: SSRF + arbitrary local file read in MiMoV2OmniMultiModalProcessor `_fetch_image` and audio loader bypass MediaConnector protections GSA_kwCzR0hTQS00aGhwLWg2NmYtajVqN84ABu8g

pypi vllm
Moderate
14 days ago

vLLM: Cross-User Data Leak Vulnerability GSA_kwCzR0hTQS03bTZoLXg5NXgtODJxNc4ABu8V

pypi vllm
Moderate
18 days ago

vLLM: Incomplete CVE-2025-62164 remediation can be bypassed by concurrent prompt parts GSA_kwCzR0hTQS1wcjdmLXA1bXctZmM4N84ABsdw

pypi vllm
Moderate
18 days ago

vLLM: ReDoS via structured_outputs.regex in the lm-format-enforcer backend (no compile timeout) — missed sibling of GHSA-rwxx-mrjm-wc2m GSA_kwCzR0hTQS00OGpoLTNnajctZmc4ds4ABsdv

pypi vllm
Moderate
18 days ago

vLLM: Unauthenticated Internal Path and Username Disclosure via Validation Error Messages GSA_kwCzR0hTQS1od3JtLWM0Y3gtcmY0as4ABsdu

pypi vllm
Moderate
18 days ago

vLLM: Derender endpoints decode caller-supplied GenerateResponse token IDs without output bounds GSA_kwCzR0hTQS04NzM3LXF4NTItaGpmZs4ABsdt

pypi vllm
Moderate
about 1 month ago

vLLM: Completion prompt lists fan out into unbounded engine requests GSA_kwCzR0hTQS04N3g1LXZtYzMtNzU2as4ABish

pypi vllm
High
2 months ago

vLLM denial of service via prompt embeds on M-RoPE models GSA_kwCzR0hTQS0zM2NnLWd4djgtM3A4Z84ABcPt

pypi vllm
Moderate
2 months ago

vLLM: Speech-to-text upload size limit is enforced after full UploadFile read GSA_kwCzR0hTQS12ODJnLTI0MzctNjdtMs4ABbGy

pypi vllm
High
2 months ago

vLLM: ReDoS via structured_outputs.regex compiled without timeout in xgrammar and outlines backends GSA_kwCzR0hTQS1yd3h4LW1yam0td2Mybc4ABbGx

pypi vllm
High
2 months ago

vLLM has Remote DoS via Invalid Recovered Token Reinjection GSA_kwCzR0hTQS04d3I1LWptMmgtOHI0Zs4ABbGw

pypi vllm
Moderate
2 months ago

vLLM: Processing differential in multi-channel audio downmixing enables hidden-input/moderation bypass for audio models GSA_kwCzR0hTQS02YzRyLWZtaDMtN3JoOM4ABbGv

pypi vllm
Moderate
3 months ago

vLLM: OOM Denial of Service via Audio Decompression Bomb GSA_kwCzR0hTQS02cHI5LXJwNTMtMnBtY84ABY4U

pypi vllm
Moderate
3 months ago

vLLM: incomplete CVE-2026-22778 fix leaks PIL repr addresses via Anthropic router GSA_kwCzR0hTQS1oZ2c4LWZxcWMtdmZtd84ABY4T

pypi vllm
Moderate
3 months ago

vLLM: GGUF dequantize kernel int truncation exposes uninitialized GPU memory in multi-tenant serving GSA_kwCzR0hTQS01anYyLWc1d3EtY21yNM4ABY4S

pypi vllm
Moderate
3 months ago

vLLM: image EXIF Rotation & PNG tRNS Transparency Not Normalized, Causing Mismatch Between Model Input and Expectations GSA_kwCzR0hTQS04anI1LXY5OHAtdzc1bc4ABY4R

pypi vllm
Moderate
3 months ago

vLLM: temperature=NaN and temperature=Infinity bypass validation and propagate to GPU kernels GSA_kwCzR0hTQS03aDRwLXJmZmctNzgyM84ABY4Q

pypi vllm
Critical
3 months ago

vLLM: OpenAI auth bypass GSA_kwCzR0hTQS05NGY0LWhyNzYtcDVqNs4ABY0L

pypi vllm
High
3 months ago

vLLM: Security Check Bypass via assert Statement in Activation Function Loading Allows Arbitrary Code Execution GSA_kwCzR0hTQS1xOGdxLTM3N3AtanEzcs4ABY0H

pypi vllm
High
3 months ago

vLLM is vulnerable to an Out-of-Memory (OOM) Denial of Service (DoS) attack due to unbounded frame count processing in the `VideoMediaIO.load_base64()` method GSA_kwCzR0hTQS13Y3dnLWM1ZmMtOXZyY84ABYjD

pypi vllm
Moderate
3 months ago

vLLM's Artifact Pin Decay allows pinned deployments to load unpinned code, weights, and processors GSA_kwCzR0hTQS0zd3c0LTVqdjktajVnbc4ABYhN

pypi vllm
Moderate
4 months ago

vllm has Improper Resource Shutdown or Release GSA_kwCzR0hTQS05OGYzLWh3ZzQtNHJmN84ABXYs

pypi vllm
Moderate
5 months ago

vLLM: extract_hidden_states speculative decoding crashes server on any request with penalty parameters GSA_kwCzR0hTQS04M3ZtLXA1MnctZjlwd84ABWWQ

pypi vllm
Moderate
5 months ago

vLLM Vulnerable to Remote DoS via Special-Token Placeholders GSA_kwCzR0hTQS1ocHY4LXgyNzYtbTU5Zs4ABWN_

pypi vllm
Low
5 months ago

vLLM makes Use of Uninitialized Resource GSA_kwCzR0hTQS14MzY4LTRnOWgtZnZ2NM4ABV1i

pypi vllm
Moderate
6 months ago

vLLM: Denial of Service via Unbounded Frame Count in video/jpeg Base64 Processing GSA_kwCzR0hTQS1wcTVjLXJqaHEtcXA3cM4ABUtx

pypi vllm
Moderate
6 months ago

vLLM: Server-Side Request Forgery (SSRF) in `download_bytes_from_url ` GSA_kwCzR0hTQS1wZjNoLXFqZ3YtdmNwcs4ABUtw

pypi vllm
Moderate
6 months ago

vLLM: Unauthenticated OOM Denial of Service via Unbounded `n` Parameter in OpenAI API Server GSA_kwCzR0hTQS0zbXdwLXd2aDktNzUyOM4ABUr7

pypi vllm
High
6 months ago

vLLM has Hardcoded Trust Override in Model Files Enables RCE Despite Explicit User Opt-Out GSA_kwCzR0hTQS03OTcyLXBnMngteHI1Oc4ABUXs

pypi vllm
Moderate
7 months ago

vLLM has SSRF Protection Bypass GSA_kwCzR0hTQS12MzU5LWpqMnYtajUzNs4ABTW2

pypi vllm
Critical
8 months ago

vLLM has RCE In Video Processing GSA_kwCzR0hTQS00cjJ4LXhwanItN2N2ds4ABRzJ

pypi vllm
High
8 months ago

vLLM vulnerable to Server-Side Request Forgery (SSRF) through MediaConnector GSA_kwCzR0hTQS1xaDRjLXhmN20tZ3hmY84ABRrE

pypi vllm
High
8 months ago

vLLM affected by RCE via auto_map dynamic module loading during model initialization GSA_kwCzR0hTQS0ycGM5LTRqODMtcWptcs4ABRWE

pypi vllm
Moderate
8 months ago

vLLM is vulnerable to DoS in Idefics3 vision models via image payload with ambiguous dimensions GSA_kwCzR0hTQS1ncmcyLTYzZnctZjJxcs4ABRBy

pypi vllm
High
9 months ago

vLLM introduced enhanced protection for CVE-2025-62164 GSA_kwCzR0hTQS1tY21jLTJtNTUtajhqas4ABQ31

pypi vllm
High
10 months ago

vLLM vulnerable to remote code execution via transformers_utils/get_config GSA_kwCzR0hTQS04ZnI0LTVxOWotbThnbc4ABPHk

pypi vllm
Moderate
10 months ago

vLLM vulnerable to DoS via large Chat Completion or Tokenization requests with specially crafted `chat_template_kwargs` GSA_kwCzR0hTQS02OWo0LWdyeGotajY0cM4ABOtt

pypi vllm
High
10 months ago

vLLM vulnerable to DoS with incorrect shape of multimodal embedding inputs GSA_kwCzR0hTQS1wbXFmLXg2eDgtcDdxd84ABOts

pypi vllm
High
10 months ago

vLLM deserialization vulnerability leading to DoS and potential RCE GSA_kwCzR0hTQS1tcnc3LWhmNGYtODNwZs4ABOtr

pypi vllm
High
12 months ago

vLLM is vulnerable to Server-Side Request Forgery (SSRF) through `MediaConnector` class GSA_kwCzR0hTQS0zZjZjLTdmdzItcHBtNM4ABNF-

pypi vllm
Moderate
12 months ago

vLLM: Resource-Exhaustion (DoS) through Malicious Jinja Template in OpenAI-Compatible Server GSA_kwCzR0hTQS02ZnZxLTIzY3ctNTYyOM4ABNF8

pypi vllm
High
12 months ago

vLLM is vulnerable to timing attack at bearer auth GSA_kwCzR0hTQS13cjloLWc3MngtbXdobc4ABNDR

pypi vllm
High
about 1 year ago

vLLM has remote code execution vulnerability in the tool call parser for Qwen3-Coder GSA_kwCzR0hTQS03OWo2LWcybTMtamdmd84ABLTx

pypi vllm
High
about 1 year ago

vllm API endpoints vulnerable to Denial of Service Attacks GSA_kwCzR0hTQS1yeGM0LTN3NnItNHY0N84ABLTs

pypi vllm
Moderate
over 1 year ago

vLLM Tool Schema allows DoS via Malformed pattern and type Fields GSA_kwCzR0hTQS12cnEzLXI4NzktN202Nc4ABIcm

pypi vllm
Moderate
over 1 year ago

vLLM allows clients to crash the openai server with invalid regex GSA_kwCzR0hTQS05aGNmLXY3bTQtNm0yas4ABIcl

pypi vllm
Moderate
over 1 year ago

vLLM DOS: Remotely kill vllm over http with invalid JSON schema GSA_kwCzR0hTQS02cWM5LXY0cjgtMjJ4Z84ABIck

pypi vllm
Moderate
over 1 year ago

vLLM has a Weakness in MultiModalHasher Image Hashing Implementation GSA_kwCzR0hTQS1jNjVwLXg2NzctZmdqNs4ABIby

pypi vllm
Low
over 1 year ago

Potential Timing Side-Channel Vulnerability in vLLM’s Chunk-Based Prefix Caching GSA_kwCzR0hTQS00cWpoLTlmdjktcjg1cs4ABIbx

pypi vllm
Moderate
over 1 year ago

vLLM vulnerable to Regular Expression Denial of Service GSA_kwCzR0hTQS1qODI4LTI4cmotaGZocM4ABIbw

pypi vllm
Moderate
over 1 year ago

vLLM has a Regular Expression Denial of Service (ReDoS, Exponential Complexity) Vulnerability in `pythonic_tool_parser.py` GSA_kwCzR0hTQS13NnE3LWo2NDItN2MyNc4ABIbv

pypi vllm
Critical
over 1 year ago

vLLM Allows Remote Code Execution via PyNcclPipe Communication Service GSA_kwCzR0hTQS1oanE0LTg3eGgtZzRmds4ABIIT

pypi vllm
High
over 1 year ago

Remote Code Execution Vulnerability in vLLM Multi-Node Cluster Configuration GSA_kwCzR0hTQS05cGNjLWd2eDUtcjV3bc4ABHiL

pypi vllm
Moderate
over 1 year ago

vLLM: Quadratic Time Complexity in Input Token Processing​ leads to denial of service GSA_kwCzR0hTQS12YzZtLWhtNDktZzlxZ84ABHQt

pypi vllm
Critical
over 1 year ago

vLLM Vulnerable to Remote Code Execution via Mooncake Integration GSA_kwCzR0hTQS1oajR3LWhtMmctcDZ3Nc4ABHQT

pypi vllm
High
over 1 year ago

Data exposure via ZeroMQ on multi-node vLLM deployment GSA_kwCzR0hTQS05ZjhmLTJ2bWYtODg1as4ABHQS

pypi vllm
Critical
over 1 year ago

CVE-2025-24357 Malicious model remote code execution fix bypass with PyTorch < 2.6.0 GSA_kwCzR0hTQS1nZ3BmLTI0anctM2Zjd84ABHGV

pypi vllm
Moderate
over 1 year ago

vLLM vulnerable to Denial of Service by abusing xgrammar cache GSA_kwCzR0hTQS1oZjNjLXd4ZzItNDlxOc4ABGyu

pypi vllm
Critical
over 1 year ago

vLLM deserialization vulnerability in vllm.distributed.GroupCoordinator.recv_object GSA_kwCzR0hTQS1wZ3I3LW1ocDUtZmdqcM4ABFtr

pypi vllm
Critical
over 1 year ago

vLLM allows Remote Code Execution by Pickle Deserialization via AsyncEngineRPCServer() RPC server entrypoints GSA_kwCzR0hTQS1jajQ3LXFqNmcteDdyNM4ABFt3

pypi vllm
Critical
over 1 year ago

vLLM Deserialization of Untrusted Data vulnerability GSA_kwCzR0hTQS01dnFyLXdwcmMtY3BwN84ABFsD

pypi vllm
Critical
over 1 year ago

vLLM Allows Remote Code Execution via Mooncake Integration GSA_kwCzR0hTQS14M204LWY3ZzUtcWhtN84ABFpg

pypi vllm
Moderate
over 1 year ago

vLLM denial of service via outlines unbounded cache on disk GSA_kwCzR0hTQS1tZ3JtLWZnanYtbWh2OM4ABFpf

pypi vllm
Low
over 1 year ago

vLLM uses Python 3.12 built-in hash() which leads to predictable hash collisions in prefix cache GSA_kwCzR0hTQS1ybTc2LTRtcmYtdjlyOM4ABEMS

pypi vllm
High
over 1 year ago

vllm: Malicious model to RCE by torch.load in hf_model_weights_iterator GSA_kwCzR0hTQS1yaDRqLTVyaHctaHI1NM4ABD24

pypi vllm
High
about 2 years ago

vLLM denial of service vulnerability GSA_kwCzR0hTQS13MnI3LTk1NzktMjdoZs4AA_m0

pypi vllm
Moderate
about 2 years ago

vLLM Denial of Service via the best_of parameter GSA_kwCzR0hTQS13YzM2LTk2OTQtZjlyZs4AA_mw

pypi vllm

Filter by Severity

Moderate 37 High 20 Critical 9 Low 3