Home/Product/vllm
Product

vllm

36 known vulnerabilities across versions
Vulnerabilities are listed by affected version. Select any CVE for the full briefing and its intelligence graph.
CVE-2026-44223
>= 0.18.0 and < 0.20.0
vLLM is an inference and serving engine for large language models (LLMs). From to before 0.20.0, the extract_hidden_states specul
6.5MEDIUM
 CVE-2026-44222
>= 0.6.1 and < 0.20.0
vLLM is an inference and serving engine for large language models (LLMs). From 0.6.1 to before 0.20.0, there is a Token Injectio
6.5MEDIUM
 CVE-2026-7141
<= 0.19.0
A vulnerability was found in vllm up to 0.19.0. The affected element is the function has_mamba_layers of the file vllm/v1/kv_cache
5.6MEDIUM
 CVE-2026-34756
>= 0.1.0 and < 0.19.0
vLLM is an inference and serving engine for large language models (LLMs). From 0.1.0 to before 0.19.0, a Denial of Service vulnera
6.5MEDIUM
 CVE-2026-34755
>= 0.7.0 and < 0.19.0
vLLM is an inference and serving engine for large language models (LLMs). From 0.7.0 to before 0.19.0, the VideoMediaIO.load_base6
6.5MEDIUM
 CVE-2026-34753
>= 0.16.0 and < 0.19.0
vLLM is an inference and serving engine for large language models (LLMs). From 0.16.0 to before 0.19.0, a server-side request forg
5.4MEDIUM
 CVE-2026-34760
>= 0.5.5 and < 0.18.0
vLLM is an inference and serving engine for large language models (LLMs). From version 0.5.5 to before version 0.18.0, Librosa def
5.9MEDIUM
 CVE-2026-27893
>= 0.10.1 and < 0.18.0
vLLM is an inference and serving engine for large language models (LLMs). Starting in version 0.10.1 and prior to version 0.18.0,
8.8HIGH
 CVE-2026-25960
>= 0.15.1 and < 0.17.0
vLLM is an inference and serving engine for large language models (LLMs). The SSRF protection fix for CVE-2026-24779 add in 0.15.1
7.1HIGH
 CVE-2026-22778
>= 0.8.3 and < 0.14.1
vLLM is an inference and serving engine for large language models (LLMs). From 0.8.3 to before 0.14.1, when an invalid image is se
9.8CRITICAL
 CVE-2026-24779
< 0.14.1
vLLM is an inference and serving engine for large language models (LLMs). Prior to version 0.14.1, a Server-Side Request Forgery (
7.1HIGH
 CVE-2026-22807
>= 0.10.1 and < 0.14.0
vLLM is an inference and serving engine for large language models (LLMs). Starting in version 0.10.1 and prior to version 0.14.0,
8.8HIGH
 CVE-2026-22773
>= 0.6.4 and < 0.12.0
vLLM is an inference and serving engine for large language models (LLMs). In versions from 0.6.4 to before 0.12.0, users can crash
6.5MEDIUM
 CVE-2025-66448
< 0.11.1
vLLM is an inference and serving engine for large language models (LLMs). Prior to 0.11.1, vllm has a critical remote code executi
7.1HIGH
 CVE-2025-62426
>= 0.5.5 and < 0.11.1
vLLM is an inference and serving engine for large language models (LLMs). From version 0.5.5 to before 0.11.1, the /v1/chat/comple
6.5MEDIUM
 CVE-2025-62372
>= 0.5.5 and < 0.11.1
vLLM is an inference and serving engine for large language models (LLMs). From version 0.5.5 to before 0.11.1, users can crash the
6.5MEDIUM
 CVE-2025-62164
>= 0.10.2 and < 0.11.1
vLLM is an inference and serving engine for large language models (LLMs). From versions 0.10.2 to before 0.11.1, a memory corrupti
8.8HIGH
 CVE-2025-59425
< 0.11.0
vLLM is an inference and serving engine for large language models (LLMs). Before version 0.11.0rc2, the API key support in vLLM pe
7.5HIGH
 CVE-2025-48956
>= 0.1.0 and < 0.10.1.1
vLLM is an inference and serving engine for large language models (LLMs). From 0.1.0 to before 0.10.1.1, a Denial of Service (DoS)
7.5HIGH
 CVE-2025-48944
>= 0.8.0 and < 0.9.0
vLLM is an inference and serving engine for large language models (LLMs). In version 0.8.0 up to but excluding 0.9.0, the vLLM bac
6.5MEDIUM
 CVE-2025-48943
>= 0.8.0 and < 0.9.0
vLLM is an inference and serving engine for large language models (LLMs). Version 0.8.0 up to but excluding 0.9.0 have a Denial of
6.5MEDIUM
 CVE-2025-48942
>= 0.8.0 and < 0.9.0
vLLM is an inference and serving engine for large language models (LLMs). In versions 0.8.0 up to but excluding 0.9.0, hitting the
6.5MEDIUM
 CVE-2025-48887
>= 0.6.4 and < 0.9.0
vLLM, an inference and serving engine for large language models (LLMs), has a Regular Expression Denial of Service (ReDoS) vulnera
6.5MEDIUM
 CVE-2025-46722
>= 0.7.0 and < 0.9.0
vLLM is an inference and serving engine for large language models (LLMs). In versions starting from 0.7.0 to before 0.9.0, in the
4.2MEDIUM
 CVE-2025-46570
< 0.9.0
vLLM is an inference and serving engine for large language models (LLMs). Prior to version 0.9.0, when a new prompt is processed,
2.6LOW
 CVE-2025-47277
>= 0.6.5 and < 0.8.5
vLLM, an inference and serving engine for large language models (LLMs), has an issue in versions 0.6.5 through 0.8.4 that ONLY imp
9.8CRITICAL
 CVE-2025-30165
>= 0.5.2
vLLM is an inference and serving engine for large language models. In a multi-node vLLM deployment using the V0 engine, vLLM uses
8.0HIGH
 CVE-2025-46560
>= 0.8.0 and < 0.8.5
vLLM is a high-throughput and memory-efficient inference and serving engine for LLMs. Versions starting from 0.8.0 and prior to 0.
6.5MEDIUM
 CVE-2025-32444
>= 0.6.5 and < 0.8.5
vLLM is a high-throughput and memory-efficient inference and serving engine for LLMs. Versions starting from 0.6.5 and prior to 0.
10.0CRITICAL
 CVE-2025-30202
>= 0.5.2 and < 0.8.5
vLLM is a high-throughput and memory-efficient inference and serving engine for LLMs. Versions starting from 0.5.2 and prior to 0.
7.5HIGH
 CVE-2024-9053
all versions
vllm-project vllm version 0.6.0 contains a vulnerability in the AsyncEngineRPCServer() RPC server entrypoints. The core functional
9.8CRITICAL
 CVE-2024-11041
all versions
vllm-project vllm version v0.6.2 contains a vulnerability in the MessageQueue.dequeue() API function. The function uses pickle.loa
9.8CRITICAL
 CVE-2025-29783
>= 0.6.5 and < 0.8.0
vLLM is a high-throughput and memory-efficient inference and serving engine for LLMs. When vLLM is configured to use Mooncake, uns
9.0CRITICAL
 CVE-2025-29770
< 0.8.0
vLLM is a high-throughput and memory-efficient inference and serving engine for LLMs. The outlines library is one of the backends
6.5MEDIUM
 CVE-2025-25183
< 0.7.2
vLLM is a high-throughput and memory-efficient inference and serving engine for LLMs. Maliciously constructed statements can lead
2.6LOW
 CVE-2025-24357
< 0.7.0
vLLM is a library for LLM inference and serving. vllm/model_executor/weight_utils.py implements hf_model_weights_iterator to load
7.5HIGH
Vulnerabilities
CISA KEV catalog
CWE weaknesses
CAPEC attack patterns
Package vulnerabilities
Threat intelligence
Threat actors
Tools & malware
ATT&CK techniques
IOCs
Detection & defense
Sigma rules
YARA rules
Atomic Red Team tests
D3FEND countermeasures
Compliance
NIST 800-53
ISO 27001:2022
SOC 2 TSC
PCI-DSS v4.0
CIS Controls v8.1
About
All capabilities
Live statistics
Data sources
Privacy policy
Terms of service
threatengine.sh  ·  Open-source threat intelligence platform  ·  100+ authoritative sources  ·  Every fact traces to its origin