ZDF
Sign inSign up

Vulnerabilities

Vulnerability · CVE-2026-114053

Arbitrary Code Injection

Published Oct 21, 2026 · catalog updated Oct 27, 2026 · Arbitrary Code Injection

Last checked Nov 4, 2026. A connected source was read within the last 7 days.

Check if you run it

No confirmed exploitation is on record for CVE-2026-114053. EPSS is 0.46%. Act only if a listed version of Arbitrary Code Injection is installed.

Gaps on this record

  • The stored product name is a package slug. The name on this page is taken from the title (Arbitrary Code Injection).
  • No vendor fix or required action is stored. Compare the linked advisory before you change production.
Severity
Critical9.2
Known exploitation
Not in CISA/VulnCheck KEV
EPSS
0.46%
Remediation
Version ranges on file
Affected products
package pip · pip vllm

## Summary An untrusted multimodal request can set . When vLLM serves a model whose Hugging Face processor uses dynamic code and the operator enabled , this request value reaches and selects the Python revision imported by Transformers. This lets a client replace the operator-reviewed processor module with code from another revision of the same model repository. The operator-selected model/data remains pinned; only the executed processor code changes. Importing a Python module is arbitrary code execution in the vLLM API/renderer process. The PoC below demonstrates this with an import-time filesystem write. It is fully offline and does not require a GPU, model weights, or changes to a real Hub repository. ## Affected snapshot - Repository: [ ](https://github.com/vllm-project/vllm) - Verified commit: [ ](https://github.com/vllm-project/vllm/commit/570735520942b6428ae98f1eb442c6d88adae55e) - Verified with Transformers 5.15.0 and 5.16.1 - Affected release range has not been bisected ## Affected configuration The vulnerable path exists when all of the following are true: 1. The served model's processor uses dynamic code. 2. The operator enabled . 3. The server can fetch the request-selected revision, or it is already cached. 4. The attacker can create or otherwise control a revision in the served model repository. An example is a [Hugging Face community pull-request ref](https://huggingface.co/docs/hub/en/repositories-pull-requests-discussions#where-in-the-git-repo-are-changes-stored) when that repository permits them. 5. A multimodal request reaches the processor path. The default configuration is not affected. ## Attack surface is accepted directly by public request models: - [ ](https://github.com/vllm-project/vllm/blob/570735520942b6428ae98f1eb442c6d88adae55e/vllm/entrypoints/serve/tokenize/protocol.py#L111-L114) - [ ](https://github.com/vllm-project/vllm/blob/570735520942b6428ae98f1eb442c6d88adae55e/vllm/entrypoints/openai/chat_completion/protocol.py#L372-L375) is p

Are you affected?

Enter the version you run. The result uses only the ranges stored on this record.

The check returns affected, not affected, or incomplete. It does not guess a version that is missing from the record.

  • package pipversion unspecified
  • pip vllm>= 0.7.3, < 0.31.0

How to fix CVE-2026-114053

What to do

Three questions, each answered from this record only. Confirm the vendor advisory before you patch.

Should you patch?

No confirmed exploitation is on record for CVE-2026-114053. EPSS is 0.46%. Act only if a listed version of Arbitrary Code Injection is installed.

Which versions are affected?

Compare what you have installed with >= 0.7.3, < 0.31.0. Compare the installed version with the linked vendor advisory before you change production.

What should an analyst verify?

CWE-829: Inclusion of Functionality from Untrusted Control Sphere; CWE-94: Code Injection. EPSS 0.46%. There is no CISA KEV listing. No Exploit-DB or Metasploit reference is linked. Use the source links for the original advisory.

Code

The project this record applies to. Exploit listings stay in References.

Timeline

Detection, updates, and resolution for this record.

  1. Oct 21, 2026 · NIST NVD

    Detected

    vLLM from 0.7.3 before 0.31.0 forwards the request-supplied mm_processor_kwargs.code_revision to Transformers AutoProcessor.from_pretrained, letting clients select which revision of a model's dynamic processor code is imported when trust_remote_code is enabled. An attacker controlling a revision in the served model repository can send a multimodal request to the unauthenticated /tokenize endpoint

  2. Oct 22, 2026 · NIST NVD

    Updated

    vLLM from 0.7.3 before 0.31.0 forwards the request-supplied mm_processor_kwargs.code_revision to Transformers AutoProcessor.from_pretrained, letting clients select which revision of a model's dynamic processor code is imported when trust_remote_code is enabled. An attacker controlling a revision in the served model repository can send a multimodal request to the unauthenticated /tokenize endpoint

  3. Oct 27, 2026 · GitHub Advisory Database

    Updated

    ## Summary An untrusted multimodal request can set . When vLLM serves a model whose Hugging Face processor uses dynamic code and the operator enabled , this request value reaches and selects the Python revision imported by Transformers. This lets a client replace the operator-reviewed processor module with code from another revision of the same model repository. The operator-selected model/data re

Sources

  • CVE.orgVulnerability recordCVE-2026-114053

    Official CVE identifier

    cve.org
  • FIRST EPSSExploitation probabilityCVE-2026-114053

    30-day exploitation probability 0.46%

    first.org
  • GitHub Advisory DatabasePackage advisoryGHSA-h3rc-6mm3-gc2m

    vLLM: Request-controlled mm_processor_kwargs.code_revision allows remote code execution

    github.com
  • NIST NVDVulnerability recordCVE-2026-114053

    vLLM from 0.7.3 before 0.31.0 forwards the request-supplied mm_processor_kwargs.code_revision to Transformers AutoProcessor.from_pretrained, letting clients select which revision of a model's dynamic processor code is imported when trust_remote_code is enabled. An attacker contro

    nvd.nist.gov
  • Snyk Vulnerability DatabasePackage advisorySNYK-PYTHON-VLLM-21347781

    Arbitrary Code Injection

    security.snyk.io

References

Freshness

Last verified in this catalog Nov 4, 2026. Significant claims (KEV, EPSS, CVSS, affected ranges) come from the source rows above — open the original publisher for the authoritative advisory.

  • GitHub Advisory DatabaseOct 28, 2026
  • Snyk Vulnerability DatabaseOct 23, 2026
  • CVE.orgOct 23, 2026
  • FIRST EPSSOct 23, 2026
  • NIST NVDOct 22, 2026