ai model report

Is pyannote/speaker-diarization-precision-2 safe to load?

HIGH risk. 3 of 87 tracked dependencies carry advisories. Driven by lightning 2.6.5 · GHSA-qqmf-gpg7-g8gw.

View on HuggingFace

// posture

fig. 01 — posture on eight axes, 100 is clean. Weakest: xyz safety at 58.

// the finding

4.2xyz score out of 10HIGH

HIGH1 confirmed finding(s) in the transitive closure; declared dependencies alone rate LOW

2 more findings on this model.

Author: pyannote
Task: voice-activity-detection
Loader: pyannote-audio
Downloads: 19,145
Remote code: not required

// dependencies

4 direct, 2 with findings, 83 reachable

Package	Evidence	Advisories	Worst
lightning	transitive, transitive	1	HIGH
torch	direct, inferred	13	none pinned
pytorch-lightning	direct, install	6	hidden
pyannote-audio	direct, install	0	hidden
torchaudio	direct, install	0	hidden
anyio	direct, install	0	hidden
asteroid-filterbanks	direct, install	0	hidden
certifi	direct, install	0	hidden

Create a free accountto see all 87 dependencies, which versions are pinned, and what to change.

// what changed

HIGHvector new2026-09-09
HIGHrisk level raised (unknown to high)2026-09-09

Risk computed 2026-09-22 at 18:58 UTC. Dependencies come from the model's requirements, its declared loader and the code it ships; advisories from NVD, GHSA and OSV.