# CVE-2026-65119: NVIDIA TensorRT-LLM untrusted tokenizer loading in KvCacheAwareRouter

Source: https://junglewise.ai/threats/cve-2026-65119-nvidia-tensorrt-llm-for-any-platform-contains-a-vulnerability-in  
Publisher: Junglewise Threat Intelligence, https://junglewise.ai/threats  
Published: 2026-10-27. Updated: 2026-10-27  
License: CC BY 4.0 with a link back

CVE-2026-65119 · Severity: high · CVSS 8.4 · Published 2026-10-27

NVIDIA TensorRT-LLM is a library for deploying and optimizing large language models. An attacker can trick the system into loading tokenizer code from an untrusted source when the KvCacheAwareRouter component processes requests. This could allow an attacker to execute arbitrary code, steal data, modify model outputs, or crash the system.

Technologies: [Nvidia TensorRT-LLM](https://junglewise.ai/threats/technologies/tensorrt-llm). Vendors: [Nvidia](https://junglewise.ai/threats/vendors/nvidia).

## Quick answers

**Is CVE-2026-65119 exploited in the wild?** No. Junglewise Threat Intelligence has no report of CVE-2026-65119 being exploited in the wild.

**How severe is CVE-2026-65119?** High severity, with a CVSS score of 8.4, published 27 October 2026.

**Which products and versions are affected?** Nvidia TensorRT-LLM unspecified.

**Is there a fix for CVE-2026-65119?** 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-2026-65119 attacked?** It is in a software library.

**Am I affected by CVE-2026-65119?** Check whether you run Nvidia TensorRT-LLM 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 KvCacheAwareRouter component in TensorRT-LLM does not properly validate the source of tokenizer functionality during loading, allowing inclusion of code from an untrusted control sphere. An attacker with the ability to influence tokenizer loading (likely requiring local or adjacent network access) can achieve code execution, information disclosure, data tampering, or denial of service depending on the deployment context.

## Affected products

- NVIDIA TensorRT-LLM unspecified

## Timeline

- 2026-10-27: disclosed

## References

- [https://github.com/NVIDIA/product-security/tree/main/2026/5883](https://github.com/NVIDIA/product-security/tree/main/2026/5883)
- [https://www.cve.org/CVERecord?id=CVE-2026-65119](https://www.cve.org/CVERecord?id=CVE-2026-65119)

## Related threats

- [NVIDIA TensorRT-LLM server-side request forgery](https://junglewise.ai/threats/cve-2026-65120-nvidia-tensorrt-llm-contains-a-vulnerability-where-an-attacker)
- [NVIDIA TensorRT-LLM server-side request forgery](https://junglewise.ai/threats/cve-2026-24269-nvidia-tensorrt-llm-contains-a-vulnerability-where-an-attacker)
- [NVIDIA TensorRT-LLM denial of service in OpenAI-compatible API](https://junglewise.ai/threats/cve-2026-47475-nvidia-tensorrt-llm-denial-of-service-in-openai-compatible-api)
- [NVIDIA TensorRT-LLM write-what-where condition](https://junglewise.ai/threats/cve-2026-47473-nvidia-tensorrt-llm-write-what-where-condition)
- [NVIDIA TensorRT-LLM deserialization in IPC layer](https://junglewise.ai/threats/cve-2026-47472-nvidia-tensorrt-llm-deserialization-in-ipc-layer)