MolGen-7: Generative Small Molecule Design

MolGen-7 is a 3.4B parameter generative model for small molecule design. Given a target and property constraints, it proposes novel candidate molecules with predicted binding affinity, ADMET scores, and synthetic accessibility.

At a glance

Field	Value
Category	Drug Discovery
Parameters	3.4B
Version	v7.0
Endpoint	`POST /v1/models/molgen-7/predict`
Output	SMILES strings with predicted affinity and property scores

Install

pip install bunker-fold

See Installation for full setup, including uv and poetry.

When to use

Generate hit candidates against a validated target.
Optimize a known scaffold for logP, molecular weight, or synthetic accessibility.
Expand a chemical library with diverse, novel structures.

Example

curl -X POST https://api.bunkerbio.com/v1/models/molgen-7/predict \
  -H "Authorization: Bearer $BUNKER_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "target": "EGFR T790M",
    "objective": "binding_affinity",
    "constraints": {
      "logP": {"max": 5},
      "mol_weight": {"max": 500},
      "synthetic_steps": {"max": 6}
    },
    "num_candidates": 4
  }'

import requests, os

r = requests.post(
    "https://api.bunkerbio.com/v1/models/molgen-7/predict",
    headers={"Authorization": f"Bearer {os.environ['BUNKER_API_KEY']}"},
    json={
        "target": "EGFR T790M",
        "objective": "binding_affinity",
        "constraints": {
            "logP": {"max": 5},
            "mol_weight": {"max": 500},
            "synthetic_steps": {"max": 6},
        },
        "num_candidates": 4,
    },
)
print(r.json())

Response

{
  "candidates_generated": 4,
  "candidates": [
    {
      "id": "BNK-4821",
      "smiles": "CC1=CC(=C...)",
      "affinity_nM": 2.1,
      "synthetic_accessibility": 3.2,
      "logP": 3.8,
      "tox_clear": true
    }
  ]
}

Parameters

string

required

Target identifier, either a gene symbol with mutation (EGFR T790M) or a UniProt ID.

string

required

Optimization objective. One of binding_affinity, selectivity, admet, or multi.

object

Property constraints applied during generation. Each key accepts min and max bounds.

integer

default:"4"

Number of candidate molecules to return. Maximum 32.

Response fields

integer

Number of candidates that passed all constraints.

array

Ranked candidate molecules.

Show candidate fields

string

Bunker Longevity-assigned candidate ID.

string

SMILES representation of the molecule.

number

Predicted binding affinity in nanomolar.

number

SA score from 1 (easy) to 10 (hard) to synthesize.

boolean

Whether the candidate clears a fast ToxPredict-v2 pre-screen.

Pipe candidates through ToxPredict-v2 for a full toxicity profile before committing to synthesis.

Get Started

Models

API Reference

Installation

MolGen-7: Generative Small Molecule Design

At a glance

Install

When to use

Example

Parameters

Response fields

​At a glance

​Install

​When to use

​Example

​Parameters

​Response fields

At a glance

Install

When to use

Example

Parameters

Response fields