Developers
Four model APIs. One key.
REST endpoints for immunotherapy, ADC, bispecific and CAR-T predictions, each returning calibrated probabilities, confidence intervals and per-feature attributions.
Endpoints
POST
/v1/models/ici/predictGAPOST
/v1/models/adc/predictEarlyPOST
/v1/models/bispecific/predictEarlyPOST
/v1/models/cart/predictDevPOST
/v1/batchGAGET
/v1/models/{id}/cardGAGET
/v1/usageGARate limits
Sandbox 10 req/min
Clinic 120 req/min
Hospital 600 req/min
SDKs
Python · TypeScript
R · FHIR bundle helper
curl https://api.cure-io.ai/v1/models/ici/predict \ -H "Authorization: Bearer $CUREIO_KEY" \ -H "Content-Type: application/json" \ -d '{ "patient": { "age": 64, "sex": "M", "tumour": "nsclc", "stage": "IV", "ecog": 1, "line": 1, "regimen": "pd1_chemo", "labs": { "nlr": 3.8, "albumin_g_l": 37, "ldh_above_uln": false }, "pdl1_tps": 55 }, "explain": true }'
from cureio import Client client = Client(api_key=os.environ["CUREIO_KEY"]) result = client.ici.predict( age=64, sex="M", tumour="nsclc", stage="IV", ecog=1, regimen="pd1_chemo", labs={"nlr": 3.8, "albumin_g_l": 37}, pdl1_tps=55, explain=True, ) print(result.response.probability) # 0.47 for f in result.response.drivers[:3]: print(f.feature, f.delta_pts)
{
"model": "ici", "version": "2.3.1",
"response": {
"probability": 0.47, "ci95": [0.39, 0.55],
"drivers": [
{ "feature": "pdl1_tps", "delta_pts": +9.1 },
{ "feature": "regimen", "delta_pts": +6.4 },
{ "feature": "nlr", "delta_pts": -3.2 }
]
},
"irae_grade3plus": {
"probability": 0.14, "ci95": [0.10, 0.19]
},
"intended_use": "decision_support",
"request_id": "req_8f2c91"
}
Get started
Bring explainable immunotherapy predictions to your team.
Book a 30-minute walkthrough for your clinical, data or pharma team, or start free in the Sandbox.