NVIDIA Triton (Python backend): Information disclosure from the Python backend

Impact

Information disclosure from the Python backend

Who can reach it

Unauthenticated network

What to do

Patch; leaks the shared-memory key that enables the write primitive

References

Related entries

• NVIDIA Triton (Python backend): Out-of-bounds write in the Python backendCVE-2025-23318 · NVIDIA Triton (Python backend)High
• vLLM (weight loading): `hf_model_weights_iterator` uses `torch.load` without `weights_only`CVE-2025-24357 · vLLM (weight loading)High
• vLLM (ZeroMQ): DoS and data exposure over ZeroMQCVE-2025-30202 · vLLM (ZeroMQ)High
• vLLM (HTTP GET): Single HTTP GET crashes the serverCVE-2025-48956 · vLLM (HTTP GET)High
• PyTorch (`torch.linalg.lu`): DoS on slice operationCVE-2025-55551 · PyTorch (`torch.linalg.lu`)High
• PyTorch (KV/conv path buffer overflow): Buffer overflow when a model combines Conv2d + hardshrink + viewCVE-2025-55558 · PyTorch (KV/conv path buffer overflow)High