Database/AI/ML frameworks & serving
PyTorch (KV/conv path buffer overflow): Buffer overflow when a model combines Conv2d + hardshrink + view
CVSS 7.5CVE-2025-55558AI/ML frameworks & servingcurated
Impact
Buffer overflow when a model combines Conv2d + hardshrink + view
Who can reach it
Customer-supplied model graph
What to do
Tenant-owned; provider isolates blast radius per container
References
Related entries
- vLLM (API key comparison): Timing attack recovers the API keyCVE-2025-59425 · vLLM (API key comparison)High
- Transformers: attacker-supplied weight-decay regex patterns hang the training process (ReDoS)CVE-2025-6921 · Hugging Face Transformers (AdamWeightDecay optimizer regex matching)High
- NVIDIA OpenShell: cleartext HTTP/2 (h2c) inside a CONNECT tunnel bypasses L7 REST policyCVE-2026-100977 · NVIDIA OpenShell for Linux (L7 REST policy enforcement)High
- NVIDIA OpenShell: CLI diagnostic command escaped for one shell layer but run through two, allowing command injectionCVE-2026-100980 · NVIDIA OpenShell for Linux (CLI diagnostic interface)High
- MLflow: schemeless prompt source lets a caller read arbitrary files readable by the tracking serverCVE-2026-116690 · MLflow model registry (_create_model_version prompt branch / get-artifact)High
- vLLM orchestrator gateway: bearer tokens and full chat payloads written to persistent logsCVE-2026-15574 · vllm-orchestrator-gateway (Red Hat OpenShift AI)High
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.