Vulnerabilities
Misconfiguration
Compliance
CVE Vulnerabilities

CVE-2026-54234

Improper Input Validation

Published: Jul 06, 2026 | Modified: Jul 07, 2026
CVSS 3.x
N/A
Source:
NVD
CVSS 2.x
RedHat/V2
RedHat/V3
7.5 IMPORTANT
CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H
Ubuntu
Vulnerability-free packages from our partners
root.io logo minimus.io logo echo.ai logo
Additional information
NVDhttps://nvd.nist.gov/vuln/detail/CVE-2026-54234
CWEhttps://cwe.mitre.org/data/definitions/20.html

vLLM is a high-throughput and memory-efficient inference and serving engine for LLMs. Prior to 0.24.0, a frontend-legal multi-request speculative decoding workload can cause the rejection sampler to produce a recovered token equal to the model vocabulary size boundary value, which is then converted to negative one when the engine selects the next live token for a request and is written back into the drafters input ids; that out-of-vocabulary value is later consumed by the models embedding and attention path and crashes the engine worker with a GPU device-side assertion. The same triggering request sequence is reachable through the public gRPC Generate and Abort endpoints, so a remote client that can send generation requests can crash the shared engine worker, aborting concurrent requests and causing a service-wide denial of service for other clients of the deployment until the worker is restarted. This issue is fixed in version 0.24.0.

Weakness

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.

Affected Software

NameVendorStart VersionEnd Version
VllmVllm*0.24.0 (excluding)
Red Hat AI Inference Server 3.2RedHatrhaiis/vllm-cuda-rhel9:1787860580*
Red Hat AI Inference Server 3.2RedHatrhaiis/vllm-rocm-rhel9:1787884873*
Red Hat AI Inference Server 3.3RedHatrhaiis/vllm-spyre-rhel9:1787161776*
Red Hat AI Inference Server 3.4RedHatrhaii/vllm-spyre-rhel9:1789681201*
Red Hat AI Inference Server 3.4RedHatrhaii/vllm-cpu-rhel9:1789681128*
Red Hat AI Inference Server 3.4RedHatrhaii/vllm-cuda-rhel9:1789681126*
Red Hat AI Inference Server 3.4RedHatrhaii/vllm-rocm-rhel9:1789681126*
Red Hat AI Inference Server 3.4RedHatrhaii/vllm-cpu-rhel9:1790075793*
Red Hat AI Inference Server 3.4RedHatrhaii/vllm-spyre-rhel9:1790076141*
Red Hat AI Inference Server 3.4RedHatrhaii/vllm-cuda-rhel9:1790090131*
Red Hat AI Inference Server 3.4RedHatrhaii/vllm-rocm-rhel9:1790109620*
Red Hat Enterprise Linux AI 3.4RedHatrhelai3/disk-image-cuda-rhel9:1790957417*
Red Hat Enterprise Linux AI 3.4RedHatrhelai3/bootc-aws-cuda-rhel9:1790881122*
Red Hat Enterprise Linux AI 3.4RedHatrhelai3/bootc-azure-cuda-rhel9:1790881120*
Red Hat Enterprise Linux AI 3.4RedHatrhelai3/bootc-azure-rocm-rhel9:1790707552*
Red Hat Enterprise Linux AI 3.4RedHatrhelai3/bootc-cuda-rhel9:1790868254*
Red Hat Enterprise Linux AI 3.4RedHatrhelai3/bootc-gcp-cuda-rhel9:1790881123*
Red Hat Enterprise Linux AI 3.4RedHatrhelai3/bootc-rocm-rhel9:1790614447*

Extended Description

Input validation is a frequently-used technique for checking potentially dangerous inputs in order to ensure that the inputs are safe for processing within the code, or when communicating with other components. Input can consist of:

Data can be simple or structured. Structured data can be composed of many nested layers, composed of combinations of metadata and raw data, with other simple or structured data. Many properties of raw data or metadata may need to be validated upon entry into the code, such as:

Implied or derived properties of data must often be calculated or inferred by the code itself. Errors in deriving properties may be considered a contributing factor to improper input validation.

Potential Mitigations

  • Assume all input is malicious. Use an “accept known good” input validation strategy, i.e., use a list of acceptable inputs that strictly conform to specifications. Reject any input that does not strictly conform to specifications, or transform it into something that does.
  • When performing input validation, consider all potentially relevant properties, including length, type of input, the full range of acceptable values, missing or extra inputs, syntax, consistency across related fields, and conformance to business rules. As an example of business rule logic, “boat” may be syntactically valid because it only contains alphanumeric characters, but it is not valid if the input is only expected to contain colors such as “red” or “blue.”
  • Do not rely exclusively on looking for malicious or malformed inputs. This is likely to miss at least one undesirable input, especially if the code’s environment changes. This can give attackers enough room to bypass the intended validation. However, denylists can be useful for detecting potential attacks or determining which inputs are so malformed that they should be rejected outright.
  • For any security checks that are performed on the client side, ensure that these checks are duplicated on the server side, in order to avoid CWE-602. Attackers can bypass the client-side checks by modifying values after the checks have been performed, or by changing the client to remove the client-side checks entirely. Then, these modified values would be submitted to the server.
  • Even though client-side checks provide minimal benefits with respect to server-side security, they are still useful. First, they can support intrusion detection. If the server receives input that should have been rejected by the client, then it may be an indication of an attack. Second, client-side error-checking can provide helpful feedback to the user about the expectations for valid input. Third, there may be a reduction in server-side processing time for accidental input errors, although this is typically a small savings.
  • Inputs should be decoded and canonicalized to the application’s current internal representation before being validated (CWE-180, CWE-181). Make sure that your application does not inadvertently decode the same input twice (CWE-174). Such errors could be used to bypass allowlist schemes by introducing dangerous inputs after they have been checked. Use libraries such as the OWASP ESAPI Canonicalization control.
  • Consider performing repeated canonicalization until your input does not change any more. This will avoid double-decoding and similar scenarios, but it might inadvertently modify inputs that are allowed to contain properly-encoded dangerous content.

References

Aqua Security is the largest pure-play cloud native security company, providing customers the freedom to innovate and run their businesses with minimal friction. The Aqua Cloud Native Security Platform provides prevention, detection, and response automation across the entire application lifecycle to secure the build, secure cloud infrastructure and secure running workloads wherever they are deployed.
Copyright © 2026 Aqua Security Software Ltd.   Privacy Policy | Terms of Use