Database/AI/ML frameworks & serving
NVIDIA Triton Inference Server (Python backend): Attacker-controlled input in the Python backend
CVSS 9.8CVE-2025-23316AI/ML frameworks & servingcurated
Impact
Attacker-controlled input in the Python backend → remote code execution
Who can reach it
Unauthenticated network to an exposed Triton HTTP/gRPC port
What to do
Patch to the fixed Triton release and rebuild NGC-derived images. Provider-published Triton images must be rebuilt, not just re-tagged
References
Related entries
- BentoML: RCE via insecure deserializationCVE-2025-27520 · BentoMLCritical
- BentoML: Insecure deserialization RCE prior to 1.4.8CVE-2025-32375 · BentoMLCritical
- PyTorch (`torch.load`): RCE via unsafe deserialization even with `weights_only=True`CVE-2025-32434 · PyTorch (`torch.load`)Critical
- vLLM (`PyNcclPipe` KV transfer): RCE via the KV cache transfer integrationCVE-2025-47277 · vLLM (`PyNcclPipe` KV transfer)Critical
- Keras: Deserialization of untrusted data in 3.11.0–3.11.2CVE-2025-49655 · KerasCritical
- Ollama (API auth): Critical authentication bypass on API endpoints through v0.12.3CVE-2025-63389 · Ollama (API auth)Critical
This entry is curated: imported from vendor advisories with machine assistance, not yet individually verified. Confirm against your vendor's advisory before acting, and report anything wrong.