# CVE-2026-47487: NVIDIA Triton Inference Server path traversal in MLflow plugin

Source: https://junglewise.ai/threats/cve-2026-47487-nvidia-triton-inference-server-path-traversal-in-mlflow-plugin  
Publisher: Junglewise Threat Intelligence, https://junglewise.ai/threats  
Published: 2026-08-04. Updated: 2026-09-25  
License: CC BY 4.0 with a link back

CVE-2026-47487 · Severity: medium · CVSS 4.4 · Published 2026-08-04

NVIDIA Triton Inference Server is a software platform that enables high-performance AI model serving. A path traversal vulnerability in the MLflow plugin allows attackers to read, write, or modify files outside the intended model repository, potentially exposing sensitive data or disrupting service availability.

Technologies: [Linux Kernel](https://junglewise.ai/threats/technologies/kernel), [Nvidia Triton Inference Server](https://junglewise.ai/threats/technologies/triton-inference-server). Vendors: [Linux](https://junglewise.ai/threats/vendors/linux), [Nvidia](https://junglewise.ai/threats/vendors/nvidia).

## Quick answers

**Is CVE-2026-47487 exploited in the wild?** No. Junglewise Threat Intelligence has no report of CVE-2026-47487 being exploited in the wild.

**How severe is CVE-2026-47487?** Medium severity, with a CVSS score of 4.4, published 4 August 2026.

**Which products and versions are affected?** Nvidia Triton Inference Server <UNKNOWN>.

**Is there a fix for CVE-2026-47487?** No fixed version is recorded yet. Check the vendor's advisory in the references below. If you run it, Junglewise drafts a mitigation that fits your setup and applies it once you approve, with senior security engineers behind it.

**How is CVE-2026-47487 attacked?** It is attacked over the network, in a hosted service.

**Am I affected by CVE-2026-47487?** Check whether you run Linux Kernel or Nvidia Triton Inference Server 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.

## Technical details

The vulnerability is a path traversal flaw in the Triton MLflow plugin that fails to properly sanitize model names provided by users. An attacker can supply a crafted path (e.g., containing "../" sequences) as a model name to access files outside the configured model repository directory. No authentication is required—the plugin accepts user-controlled input directly. Successful exploitation allows reading arbitrary files (information disclosure), writing or modifying files outside the repository, and potentially causing denial of service. A patch is expected from NVIDIA.

## Affected products

- NVIDIA Triton Inference Server <UNKNOWN>

## Timeline

- 2026-08-04: disclosed

## References

- [https://github.com/NVIDIA/product-security/tree/main/2026/5860](https://github.com/NVIDIA/product-security/tree/main/2026/5860)
- [https://www.cve.org/CVERecord?id=CVE-2026-47487](https://www.cve.org/CVERecord?id=CVE-2026-47487)

## Related threats

- [Linux IOMMU page table logic error in arm_lpae_split_blk_unmap](https://junglewise.ai/threats/cve-2026-56934-in-arm-lpae-split-blk-unmap-of-drivers-iommu-io-pgtable-arm)
- [Linux kernel null pointer dereference in TCP established lookup](https://junglewise.ai/threats/cve-2026-99288-in-the-linux-kernel-the-following-vulnerability-has-been-resolved)
- [Linux kernel xen-netback race condition in XenBus disconnect](https://junglewise.ai/threats/cve-2026-99286-in-the-linux-kernel-the-following-vulnerability-has-been-resolved)
- [Linux kernel shared-frag flag loss in skb_zerocopy](https://junglewise.ai/threats/cve-2026-99284-in-the-linux-kernel-the-following-vulnerability-has-been-resolved)
- [Linux kernel netfilter conntrack TCP RST timeout handling](https://junglewise.ai/threats/cve-2026-99283-in-the-linux-kernel-the-following-vulnerability-has-been-resolved)