NVIDIA TensorRT-LLM Triton Backend context_phase_params Deserialization Enables SSRF + RDMA Injection

Affected Vendor(s)

Affected Product(s)

Summary

Mindgard discovered a vulnerability in NVIDIA TensorRT-LLM (v1.3.0rc5), a distributed large language model inference framework for GPU-accelerated serving. The issue affects TensorRT-LLM’s disaggregated serving architecture and manifests in how serialized communication state is transferred between inference components.

Timeline

Discovered on
February 23, 2026
Disclosed to Vendor on
February 25, 2026
Published on
November 10, 2026

Credit

Blog Post

References