ADMET-Guard
ADMET-Guard is a SYNTHETIC/DEMO drug discovery model for admet property + toxicity prediction. Outputs are deterministic demo values — NOT for clinical use. Trained (synthetically) on de-identified tabular data.
Tabular
BioNexus
License: BioFoundry Research
v1.5.4
RTG 3.8
DL 301,640
Metrics
| Metric | Value | Split |
|---|---|---|
| ROC-AUC | 0.916 | Synthetic hold-out |
| Calibration (ECE) | 0.074 | Synthetic |
| Inputs | Outputs |
|---|---|
| Molecular descriptors | ADMET panel + tox flags |
Run a synthetic prediction
Returns a deterministic synthetic result — clearly labeled, not for clinical use.
API integration
Call this model from your pipeline (demo — no real key required):
import requests
r = requests.post("https://api.biofoundry.ai/v1/models/bf-015-admet-guard/predict",
headers={"Authorization": "Bearer <API_KEY>"},
json={"input": "<molecular descriptors>"})
print(r.json()) # -> synthetic demo prediction
See the API reference · JSON: /api/models/bf-015-admet-guard
Version history
| Version | Status | Notes |
|---|---|---|
| v1.5.0 | Archived | Synthetic release note 0 |
| v1.5.1 | Archived | Synthetic release note 1 |
| v1.5.2 | Archived | Synthetic release note 2 |
| v1.5.3 | Archived | Synthetic release note 3 |
| v1.5.4 | Current | Synthetic release note 4 |
Community contributions: 73 synthetic contributors have submitted evaluations and fine-tunes.