# CVE-2026-27893: vLLM remote code execution via hardcoded remote code trust in models

Source: https://junglewise.ai/threats/cve-2026-27893-vllm-remote-code-execution-via-hardcoded-remote-code-trust-in  
Publisher: Junglewise Threat Intelligence, https://junglewise.ai/threats  
Published: 2026-03-27. Updated: 2026-07-17  
License: CC BY 4.0 with a link back

CVE-2026-27893 · Severity: high · CVSS 8.8 · Published 2026-03-27

vLLM is a high-performance engine used to run large language models (LLMs). A security flaw was found where the software ignores a user's request to block untrusted code when loading certain specific AI models (Nemotron and Kimi). If an attacker convinces a user to load a malicious model from a remote repository, they could execute arbitrary commands on the user's system, potentially leading to full system takeover or data theft.

Technologies: [vllm (PyPI)](https://junglewise.ai/threats/technologies/pypi-vllm), [Red Hat AI Inference Server](https://junglewise.ai/threats/technologies/ai-inference-server), [Vllm](https://junglewise.ai/threats/technologies/vllm). Vendors: [PyPI](https://junglewise.ai/threats/vendors/pypi), [Red Hat](https://junglewise.ai/threats/vendors/red-hat), [Vllm](https://junglewise.ai/threats/vendors/vllm).

## Quick answers

**Is CVE-2026-27893 exploited in the wild?** No. Junglewise Threat Intelligence has no report of CVE-2026-27893 being exploited in the wild.

**How severe is CVE-2026-27893?** High severity, with a CVSS score of 8.8, published 27 March 2026.

**Which products and versions are affected?** vllm (PyPI) ECOSYSTEM:introduced=0.10.1,fixed=0.18.0; Red Hat AI Inference Server 3.2, 3.3.

**Is there a fix for CVE-2026-27893?** Yes. vllm (PyPI): upgrade to 0.18.0 or later. Junglewise finds every place you run it, drafts the upgrade and applies it once you approve.

**How is CVE-2026-27893 attacked?** It is attacked over the network, in a software library.

**Am I affected by CVE-2026-27893?** Check whether you run vllm (PyPI), Red Hat AI Inference Server or Vllm at an affected version. Junglewise does that check for you around the clock: connect your code, cloud and domains, and it matches new vulnerabilities like this one against what you actually run and tells you when one reaches you.

## How to fix

- vllm (PyPI): upgrade to 0.18.0 or later

  ```sh
  pip install "vllm>=0.18.0"
  ```


## Technical details

A protection mechanism failure (CWE-693) exists in vLLM's model executor. Specifically, the implementation files for Nemotron-VL and Kimi-K25 hardcode 'trust_remote_code=True' during calls to HuggingFace API components (AutoModel.from_config and cached_get_image_processor). This hardcoded value overrides the global user-defined security setting intended to prevent the execution of untrusted remote code. An attacker can exploit this by hosting a malicious model repository that, when loaded by a vulnerable vLLM instance, executes arbitrary Python code. The vulnerability was patched in version 0.18.0 by ensuring these components respect the global model configuration.

## Affected products

- vllm-project vllm >= 0.10.1, < 0.18.0

## Identifiers

- CVE-2026-27893
- GHSA-7972-pg2x-xr59
- PYSEC-2026-2297

## Timeline

- 2026-03-26: disclosed
- 2026-03-27: advisory
- 2026-03-27: patched: Version 0.18.0 released

## References

- [https://api.github.com/users/Wernerina](https://api.github.com/users/Wernerina)
- [https://github.com/Wernerina](https://github.com/Wernerina)
- [https://api.github.com/users/Wernerina/gists%7B/gist_id%7D](https://api.github.com/users/Wernerina/gists%7B/gist_id%7D)
- [https://api.github.com/users/Wernerina/repos](https://api.github.com/users/Wernerina/repos)
- [https://avatars.githubusercontent.com/u/151374244?v=4](https://avatars.githubusercontent.com/u/151374244?v=4)
- [https://api.github.com/users/Wernerina/events%7B/privacy%7D](https://api.github.com/users/Wernerina/events%7B/privacy%7D)

## Related threats

- [vLLM chat template resource exhaustion DoS](https://junglewise.ai/threats/cve-2026-115918-vllm-is-an-inference-and-serving-engine-for-large-language)
- [vLLM: Permissive default CORS origins allow cross-origin requests to the vLLM HTTP server](https://junglewise.ai/threats/cve-2026-115917-vllm-is-an-inference-and-serving-engine-for-large-language)
- [vLLM off-by-one error in tokenizer max_token_id validation](https://junglewise.ai/threats/cve-2026-115916-vllm-is-an-inference-and-serving-engine-for-large-language)
- [vLLM image format validation bypass in media processing](https://junglewise.ai/threats/cve-2026-115915-vllm-is-an-inference-and-serving-engine-for-large-language)
- [vLLM denial of service in DeepSeek-OCR image processing](https://junglewise.ai/threats/cve-2026-115914-vllm-is-an-inference-and-serving-engine-for-large-language)