Lucene search
+L

111 matches found

RedhatCVE
RedhatCVE
•added 2026/10/21 5:52 p.m.•7 views

CVE-2026-114052

A flaw was found in vLLM. This vulnerability allows an unauthenticated remote attacker to cause a Denial of Service DoS via memory exhaustion. When serving MOSS-Audio models, the unauthenticated tokenize endpoint caches processor configurations in an unbounded cache. By submitting requests with...

8.7CVSS5.8AI score0.00479EPSS
SaveExploits0References5
Snyk
Snyk
•added 2026/10/21 3:00 a.m.•10 views

Arbitrary Code Injection

Overview vllm is an A high-throughput and memory-efficient inference and serving engine for LLMs Affected versions of this package are vulnerable to Arbitrary Code Injection via the /tokenize API endpoint, which does not enforce authentication even when the --api-key option is configured. An...

9.2CVSS5.8AI score0.00462EPSS
SaveExploits0References2
Snyk
Snyk
•added 2026/10/21 3:00 a.m.•10 views

Allocation of Resources Without Limits or Throttling

Overview vllm is an A high-throughput and memory-efficient inference and serving engine for LLMs Affected versions of this package are vulnerable to Allocation of Resources Without Limits or Throttling via MossAudioProcessingInfo.gethfprocessor in MOSS-Audio model serving, which caches a...

8.7CVSS5.8AI score0.00479EPSS
SaveExploits0References2
NVD
NVD
•added 2026/10/21 1:17 a.m.•9 views

CVE-2026-114053

vLLM from 0.7.3 before 0.31.0 forwards the request-supplied mmprocessorkwargs.coderevision to Transformers AutoProcessor.frompretrained, letting clients select which revision of a model's dynamic processor code is imported when trustremotecode is enabled. An attacker controlling a revision in the...

9.2CVSS0.00462EPSS
SaveExploits0References3
NVD
NVD
•added 2026/10/21 1:17 a.m.•7 views

CVE-2026-114052

vLLM after 0.24.0 and before 0.31.0 contains a memory exhaustion vulnerability in MOSS-Audio model serving that caches a processor for every distinct request-controlled mmprocessorkwargs value in an unbounded dictionary. Unauthenticated attackers can send distinct or oversized melconfig values to...

8.7CVSS0.00479EPSS
SaveExploits0References2
EUVD
EUVD
•added 2026/10/21 12:36 a.m.•13 views

EUVD-2026-107003

vLLM after 0.24.0 and before 0.31.0 contains a memory exhaustion vulnerability in MOSS-Audio model serving that caches a processor for every distinct request-controlled mmprocessorkwargs value in an unbounded dictionary. Unauthenticated attackers can send distinct or oversized melconfig values to...

8.7CVSS6AI score0.00479EPSS
SaveExploits0References2
CVE
CVE
•added 2026/10/21 12:36 a.m.•17 views

CVE-2026-114052

The vLLM framework (versions after 0.24.0 and before 0.31.0) contains a memory exhaustion vulnerability within the MOSS-Audio model serving component. The root cause is the use of an unbounded dictionary that caches a processor for every unique request-controlled mm_processor_kwargs value. Unauth...

8.7CVSS5.9AI score0.00479EPSS
SaveExploits0References2
osv_cve
osv_cve
•added 2026/10/21 12:36 a.m.•7 views

CVE-2026-114053 vLLM 0.7.3 before 0.31.0 RCE via mm_processor_kwargs.code_revision

vLLM from 0.7.3 before 0.31.0 forwards the request-supplied mmprocessorkwargs.coderevision to Transformers AutoProcessor.frompretrained, letting clients select which revision of a model's dynamic processor code is imported when trustremotecode is enabled. An attacker controlling a revision in the...

9.2CVSS6.7AI score
SaveExploits0References5
EUVD
EUVD
•added 2026/10/21 12:36 a.m.•13 views

EUVD-2026-107004

vLLM from 0.7.3 before 0.31.0 forwards the request-supplied mmprocessorkwargs.coderevision to Transformers AutoProcessor.frompretrained, letting clients select which revision of a model's dynamic processor code is imported when trustremotecode is enabled. An attacker controlling a revision in the...

9.2CVSS6.7AI score0.00462EPSS
SaveExploits0References3
osv_cve
osv_cve
•added 2026/10/21 12:36 a.m.•10 views

CVE-2026-114052 vLLM 0.25.0 before 0.31.0 MOSS-Audio Memory Exhaustion via /tokenize mm_processor_kwargs

vLLM after 0.24.0 and before 0.31.0 contains a memory exhaustion vulnerability in MOSS-Audio model serving that caches a processor for every distinct request-controlled mmprocessorkwargs value in an unbounded dictionary. Unauthenticated attackers can send distinct or oversized melconfig values to...

8.7CVSS6AI score
SaveExploits0References4
Mend
Mend
•added 2026/10/21 12:36 a.m.•9 views

CVE-2026-114052

vLLM after 0.24.0 and before 0.31.0 contains a memory exhaustion vulnerability in MOSS-Audio model serving that caches a processor for every distinct request-controlled mmprocessorkwargs value in an unbounded dictionary. Unauthenticated attackers can send distinct or oversized melconfig values to...

8.7CVSS5.9AI score0.00479EPSS
SaveExploits0References2
Cvelist
Cvelist
•added 2026/10/21 12:36 a.m.•24 views

CVE-2026-114052 vLLM 0.25.0 before 0.31.0 MOSS-Audio Memory Exhaustion via /tokenize mm_processor_kwargs

vLLM after 0.24.0 and before 0.31.0 contains a memory exhaustion vulnerability in MOSS-Audio model serving that caches a processor for every distinct request-controlled mmprocessorkwargs value in an unbounded dictionary. Unauthenticated attackers can send distinct or oversized melconfig values to...

8.7CVSS0.00479EPSS
SaveExploits0References2
Vulnrichment
Vulnrichment
•added 2026/10/21 12:36 a.m.•6 views

CVE-2026-114052 vLLM 0.25.0 before 0.31.0 MOSS-Audio Memory Exhaustion via /tokenize mm_processor_kwargs

vLLM after 0.24.0 and before 0.31.0 contains a memory exhaustion vulnerability in MOSS-Audio model serving that caches a processor for every distinct request-controlled mmprocessorkwargs value in an unbounded dictionary. Unauthenticated attackers can send distinct or oversized melconfig values to...

8.7CVSS5.9AI score0.00479EPSS
SaveExploits0References2
Cvelist
Cvelist
•added 2026/10/21 12:36 a.m.•24 views

CVE-2026-114053 vLLM 0.7.3 before 0.31.0 RCE via mm_processor_kwargs.code_revision

vLLM from 0.7.3 before 0.31.0 forwards the request-supplied mmprocessorkwargs.coderevision to Transformers AutoProcessor.frompretrained, letting clients select which revision of a model's dynamic processor code is imported when trustremotecode is enabled. An attacker controlling a revision in the...

9.2CVSS0.00462EPSS
SaveExploits0References3
Mend
Mend
•added 2026/10/21 12:36 a.m.•13 views

CVE-2026-114053

vLLM from 0.7.3 before 0.31.0 forwards the request-supplied mmprocessorkwargs.coderevision to Transformers AutoProcessor.frompretrained, letting clients select which revision of a model's dynamic processor code is imported when trustremotecode is enabled. An attacker controlling a revision in the...

9.2CVSS6.7AI score0.00462EPSS
SaveExploits0References3
CVE
CVE
•added 2026/10/21 12:36 a.m.•34 views

CVE-2026-114053

The vLLM library versions 0.7.3 before 0.31.0 is vulnerable to Remote Code Execution (RCE) via the mm_processor_kwargs.code_revision parameter. The application forwards this user-supplied revision to Transformers AutoProcessor.from_pretrained . If trust_remote_code is enabled, an attacker who con...

9.2CVSS6.7AI score0.00462EPSS
SaveExploits0References3
Positive Technologies
Positive Technologies
•added 2026/10/21 12:00 a.m.•13 views

PT-2026-120179

Name of the Vulnerable Software and Affected Versions vLLM versions 0.7.3 through 0.30.x Description When trust remote code is enabled, the software forwards the request-supplied mm processor kwargs.code revision to the Transformers AutoProcessor.from pretrained function. This allows clients to...

9.2CVSS6.6AI score0.00462EPSS
SaveExploits0References8
Positive Technologies
Positive Technologies
•added 2026/10/21 12:00 a.m.•9 views

PT-2026-120178

vLLM after 0.24.0 and before 0.31.0 contains a memory exhaustion vulnerability in MOSS-Audio model serving that caches a processor for every distinct request-controlled mm processor kwargs value in an unbounded dictionary. Unauthenticated attackers can send distinct or oversized mel config values...

8.7CVSS6AI score0.00479EPSS
SaveExploits0References3
EUVD
EUVD
•added 2026/10/05 11:42 p.m.•17 views

EUVD-2026-92848

vLLM: Qwen2-VL / Qwen3-VL video samplers bound on request-controlled maxframes, which the numframes ceiling does not reach...

5.3CVSS5.9AI score0.00303EPSS
SaveExploits0References4
Github Security Blog
Github Security Blog
•added 2026/10/05 11:42 p.m.•28 views

vLLM: Qwen2-VL / Qwen3-VL video samplers bound on request-controlled max_frames, which the num_frames ceiling does not reach

Summary An unauthenticated remote attacker can exhaust the memory of the vLLM API-server process by raising the request-level mediaiokwargs.video.maxframes and fps knobs on any deployment serving a Qwen2-VL or Qwen3-VL model. 74 extra bytes of JSON took the server's peak RSS from 2 271 MiB to 13...

5.3CVSS6AI score0.00303EPSS
SaveExploits0References5Affected Software1
Rows per page
Query Builder