GenerationGPU8 credits

REINVENT4 — Molecule Generation

AI-powered de novo drug design and lead optimization

REINVENT4 generates novel drug-like molecules using reinforcement learning with multi-objective optimization. Supports de novo design, scaffold hopping, R-group replacement, and linker design. Developed by AstraZeneca's Molecular AI team.

$0.08

per API call

credits per call

/v1/generate/molecules

API endpoint

Features

De novo molecule design

Scaffold hopping between chemical series

Linker design for PROTACs and molecular glues

R-group replacement and enumeration

Multi-objective optimization

Reinforcement learning with customizable scoring

Quick Start

REINVENT4 — Python Examplepython

import requests

API_KEY = "sk-sci-your-key-here"
url = "https://scirouter.ai/v1/generate/molecules"

response = requests.post(url, json={
    "mode": "de_novo",
    "scoring_functions": ["QED", "SA_score"],
    "num_molecules": 50,
    "constraints": {
        "molecular_weight": {"min": 300, "max": 500},
        "logp": {"min": 1, "max": 4}
    }
}, headers={"Authorization": f"Bearer {API_KEY}"})

result = response.json()
for mol in result["molecules"][:5]:
    print(f"SMILES: {mol['smiles']}")
    print(f"  QED: {mol['qed']:.3f} | SA: {mol['sa_score']:.2f}")