# CVE-2026-116177: vLLM uncontrolled resource consumption in Nemotron-VL processor

Source: https://junglewise.ai/threats/cve-2026-116177-vllm-from-0-10-2-before-0-25-0-contains-an-uncontrolled-resource  
Publisher: Junglewise Threat Intelligence, https://junglewise.ai/threats  
Published: 2026-10-25. Updated: 2026-10-25  
License: CC BY 4.0 with a link back

CVE-2026-116177 · Severity: high · CVSS 7.5 · Published 2026-10-25

vLLM is an open-source framework for serving large language models. Versions 0.10.2 through 0.24.x contain a vulnerability where the Nemotron-VL image processor disables safety guards in the image parsing library, allowing unauthenticated attackers to submit specially crafted images that consume excessive memory. This can crash the vLLM service and deny it to other users.

Technologies: [vllm (PyPI)](https://junglewise.ai/threats/technologies/pypi-vllm), [vLLM Project vllm](https://junglewise.ai/threats/technologies/vllm-project-vllm). Vendors: [PyPI](https://junglewise.ai/threats/vendors/pypi), [vLLM Project](https://junglewise.ai/threats/vendors/vllm-project).

## Quick answers

**Is CVE-2026-116177 exploited in the wild?** No. Junglewise Threat Intelligence has no report of CVE-2026-116177 being exploited in the wild.

**How severe is CVE-2026-116177?** High severity, with a CVSS score of 7.5, published 25 October 2026.

**Which products and versions are affected?** vllm (PyPI) ECOSYSTEM:introduced=0.10.2,fixed=0.25.0; vLLM Project vllm 0.10.2 before 0.25.0.

**Is there a fix for CVE-2026-116177?** Yes. vllm (PyPI): upgrade to 0.25.0 or later. Until then: Set the environment variable VLLM_MAX_IMAGE_PIXELS to restrict the maximum decoded pixel count for images, or disable Nemotron-VL model serving until upgrading. Junglewise finds every place you run it, drafts the upgrade and applies it once you approve.

**How is CVE-2026-116177 attacked?** It is attacked over the network.

**Am I affected by CVE-2026-116177?** Check whether you run vllm (PyPI) or vLLM Project 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 0.25.0 or later

  ```sh
  pip install "vllm>=0.25.0"
  ```


Set the environment variable VLLM_MAX_IMAGE_PIXELS to restrict the maximum decoded pixel count for images, or disable Nemotron-VL model serving until upgrading.

## Technical details

The Nemotron-VL processors in vllm/transformers_utils/processors/nemotron_vl.py and nano_nemotron_vl.py disable Pillow's decompression-bomb defence by setting Image.MAX_IMAGE_PIXELS = None at module import time. An unauthenticated attacker reaching /v1/chat/completions can submit a pathological image that decompresses into substantial memory, triggering out-of-memory conditions. The process-global setting affects all image loading in that vLLM worker process, potentially impacting unrelated operations (CWE-400).

## Affected products

- vLLM Project vLLM 0.10.2 before 0.25.0

## Identifiers

- CVE-2026-116177
- GHSA-mm38-5g96-7j6x
- PYSEC-2026-4352

## Timeline

- 2026-10-09: disclosed: GitHub security advisory published (GHSA-mm38-5g96-7j6x)
- 2026-10-25: advisory: CVE-2026-116177 published

## References

- [https://github.com/vllm-project/vllm/security/advisories/GHSA-mm38-5g96-7j6x](https://github.com/vllm-project/vllm/security/advisories/GHSA-mm38-5g96-7j6x)
- [https://www.vulncheck.com/advisories/vllm-0.10.2-before-0.25.0-dos-via-nemotron-vl-processor-decompression-bomb-guard-bypass](https://www.vulncheck.com/advisories/vllm-0.10.2-before-0.25.0-dos-via-nemotron-vl-processor-decompression-bomb-guard-bypass)

## 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)