Database/AI/ML frameworks & serving
Ollama (GGUF model loader): Heap out-of-bounds read from an attacker-supplied GGUF via `/api/create`
CVSS 9.1CVE-2026-7482AI/ML frameworks & servingcurated
Impact
Heap out-of-bounds read from an attacker-supplied GGUF via /api/create
Who can reach it
Customer-supplied model file to an exposed API
What to do
Upgrade to 0.17.1+
References
Related entries
- Cocos AI: TDX aTLS verifier ignores expected REPORT_DATA, accepting replayed attestation evidenceCVE-2026-92701 · Cocos AI (Intel TDX aTLS attestation verifier, VerifTeeAttestation)Critical
- Cocos AI: SEV-SNP verifier skips ReportData binding when caller passes empty expected valueCVE-2026-92702 · Cocos AI (AMD SEV-SNP attestation verifier, VerifyAttestationReportTLS)Critical
- TensorFlow (SavedModel protobuf): Mutating a SavedModel protobuf crashes or corrupts the serving processCVE-2020-15206 · TensorFlow (SavedModel protobuf)Critical
- Jupyter Server Proxy: Authentication weakness in proxied-process accessCVE-2024-28179 · Jupyter Server ProxyCritical
- vLLM (Mooncake): Unsafe deserialization over ZMQ/TCP bound to all interfacesCVE-2025-29783 · vLLM (Mooncake)Critical
- Flowise: custom-function sandbox escape gives authenticated users OS command executionCVE-2026-119978 · Flowise (/api/v1/node-custom-function, vm2 sandbox with ALLOW_BUILTIN_DEP=true)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.