CVE-2025-48944VulnerabilityPublished 30 May 2025

CVE-2025-48944: vllm · Improper Input Validation

Affects vllmSource CVE Database V5

Not known to be exploited · CVSS 6.5 Medium : upgrade to 0.9.0 when convenient.

Data as of 26 Oct 2026, 06:26 UTC

CVSS 3.1 base score

6.5Medium

How severe the flaw is on the 0–10 CVSS scale.

EPSS: 0.6% chance of exploitation in 30 days – higher than 49% of CVEsExploit Prediction Scoring System: the likelihood attackers will use this flaw soon.
Patch: Available – fixed in 0.9.0

CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:H

Attack Vector: Network
Attack Complexity: Low
Privileges Required: Low
User Interaction: None
Scope: Unchanged
Confidentiality: None
Integrity: None
Availability: High

Bold values are the worst case: easiest to exploit or largest impact.

Attack Vector: Network: The vulnerable component is bound to the network stack and the set of possible attackers extends beyond the other options listed, up to and including the entire Internet. Such a vulnerability is often termed 'remotely exploitable' and can be thought of as an attack being exploitable at the protocol level one or more network hops away (e.g., across one or more routers).
Attack Complexity: Low: Specialized access conditions or extenuating circumstances do not exist. An attacker can expect repeatable success against the vulnerable component.
Privileges Required: Low: The attacker is authorized with (i.e., requires) privileges that provide basic user capabilities that could normally affect only settings and files owned by a user. Alternatively, an attacker with Low privileges may have the ability to cause an impact only to non-sensitive resources.
User Interaction: None: The vulnerable system can be exploited without any interaction from any user.
Scope: Unchanged: An exploited vulnerability can only affect resources managed by the same security authority. In this case, the vulnerable component and the impacted component are either the same, or both are managed by the same security authority.
Confidentiality: None: There is no loss of confidentiality within the impacted component.
Integrity: None: There is no loss of integrity within the impacted component.
Availability: High: There is total loss of availability, resulting in the attacker being able to fully deny access to resources in the impacted component; this loss is either sustained (while the attacker continues to deliver the attack) or persistent (the condition persists even after the attack has completed). Alternatively, the attacker has the ability to deny some availability, but the loss of availability presents a direct, serious consequence to the impacted component (e.g., the attacker cannot disrupt existing connections, but can prevent new connections; the attacker can repeatedly exploit a vulnerability that, in each instance of a successful attack, leaks a only small amount of memory, but after repeated exploitation causes a service to become completely unavailable).

Definitions: FIRST CVSS v3.1 specification.

Exploitation evidence

Disclosed: evidence found.
CVE published 30 May 2025
Detectable: no evidence on record.
No public nuclei template indexed
Weaponised: no evidence on record.
No public exploit code indexed in Metasploit or Exploit-DB
Exploited in the wildExploited: no evidence on record.
Not in CISA KEV and no source reports exploitation
Observed by OffSeq MirageSeen by Mirage: no evidence on record.
No attack traffic recorded by OffSeq Mirage sensors

A missing step is not evidence of safety. Sources: cve.org (opens in a new tab), the Metasploit, Exploit-DB and nuclei-templates index as of 2026-10-26.

Disclosed: The CVE record is public on cve.org. Its publication is t = 0.
Detectable: A public nuclei template can find exposed instances, for defenders and attackers alike.
Weaponised: Public exploit code exists in Metasploit or Exploit-DB, or CISA SSVC reports a public proof of concept.
Exploited in the wild: CISA KEV lists it, CISA SSVC rates exploitation active, or a source reports attacks.
Observed by OffSeq Mirage: OffSeq Mirage honeypot sensors recorded attack traffic for this CVE, or found exposed hosts without attack traffic.

Moments are spaced evenly, not to scale; a dashed line marks a gap of six months or more. Deltas count from t = 0. Sources that give only a date (CISA KEV, Exploit-DB, or a time of exactly 00:00Z) read “same day” rather than hours. Metasploit modules carry the flaw’s disclosure date, not their release date, and nuclei templates carry no date, so both appear above but not on the timeline.

What to do

Upgrade vllm-project/vllm to 0.9.0.

Affected software

Product	Affected	Fixed in
GitHubvllm-project/vllm`pkg:github/vllm-project/vllm`Other threats	Affected <0.9.0 >=0.8.0	Fixed in 0.9.0

Check your own hosts for these packages with threat-finder (opens in a new tab), our free open-source scanner.

AI-suggested

European organizations should prioritize upgrading vLLM to version 0.9.0 or later, where this input validation flaw is fixed. Until the upgrade is applied, organizations should implement strict input validation and sanitization at the API gateway or proxy level to filter out malformed or unexpected inputs in the "pattern" and "type" fields. Rate limiting and anomaly detection on the /v1/chat/completions endpoint can help detect and block suspicious requests that may trigger crashes. Monitoring the health of inference workers and automating restart procedures can reduce downtime impact. Additionally, restricting access to the API endpoint to trusted networks or authenticated users with minimal privileges can reduce exploitation risk. Organizations should also review internal policies to limit who can invoke the tools functionality and audit logs for unusual request patterns. Finally, security teams should stay alert for any emerging exploit reports and apply patches promptly.

Description

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 backend used with the /v1/chat/completions OpenAPI endpoint fails to validate unexpected or malformed input in the "pattern" and "type" fields when the tools functionality is invoked. These inputs are not validated before being compiled or parsed, causing a crash of the inference worker with a single request. The worker will remain down until it is restarted. Version 0.9.0 fixes the issue.

Impact

For European organizations deploying vLLM versions 0.8.0 to <0.9.0, this vulnerability poses a significant availability risk to AI inference services. Organizations relying on AI-driven chat completions or other LLM-based tools for customer support, automation, or decision-making could experience service outages due to worker crashes triggered by malicious or malformed API requests. This disruption could degrade user experience, reduce operational efficiency, and potentially impact business continuity. While the vulnerability does not expose sensitive data or allow unauthorized code execution, the denial of service could be exploited by attackers to cause repeated outages or degrade trust in AI services. Given the growing adoption of AI technologies in Europe, especially in sectors like finance, healthcare, and public services, the impact could be notable if the vulnerable versions are in use. Additionally, the requirement for low privileges means that insider threats or compromised internal systems could exploit this vulnerability to disrupt AI services.

CWE-20Improper Input Validation (opens in a new tab)
Class weaknessLikelihood of exploit: HighCWE Top 25 (2025): #18OWASP Top 10 A05:2025: Injection (opens in a new tab)
The product receives input or data, but it does not validate or incorrectly validates that the input has the properties that are required to process the data safely and correctly.
MITRE discourages mapping vulnerabilities to this broad entry; the root cause is usually a more specific weakness.
Architecture and Design · Attack Surface Reduction
Consider using language-theoretic security (LangSec) techniques that characterize inputs using a formal language and build "recognizers" for that language. This effectively requires parsing to be a distinct layer that effectively enforces a boundary between raw input and internal data representations, instead of…
Architecture and Design · Libraries or Frameworks
Use an input validation framework such as Struts or the OWASP ESAPI Validation API. Note that using a framework does not automatically address all input validation problems; be mindful of weaknesses that could arise from misusing the framework itself (CWE-1173).
Architecture and Design · Implementation · Attack Surface Reduction
Understand all the potential areas where untrusted inputs can enter the product, including but not limited to: parameters or arguments, cookies, anything read from the network, environment variables, reverse DNS lookups, query results, request headers, URL components, e-mail, files, filenames, databases, and any…
OWASP Cheat Sheets
XML Security (opens in a new tab)
Mass Assignment (opens in a new tab)
Injection Prevention (opens in a new tab)
SQL Injection Prevention (opens in a new tab)
Injection Prevention in Java (opens in a new tab)
OWASP ASVS 4.0.3
V13.3.1 (opens in a new tab)
V5.1.4 (opens in a new tab)
V5.1.3 (opens in a new tab)
Attack patterns
Buffer Overflow via Environment Variables (opens in a new tab)
Server Side Include (SSI) Injection (opens in a new tab)
Cross Zone Scripting (opens in a new tab)
Command Line Execution through SQL Injection (opens in a new tab)
Object Relational Mapping Injection (opens in a new tab)
From MITRE CWE, MITRE CAPEC and OWASP (Top 10, OpenCRE).

References

Community reviews

Loading reviews…

Exploitation evidence

What to do

Affected software

Description

AI analysisMachine-generated, updated 8 Jul 2025, 14:25 UTC

WeaknessesCWE-20

Technical details

References

Community reviews

Related threats