# chemprop/chemprop

Message Passing Neural Networks for Molecule Property Prediction

Repository: https://github.com/chemprop/chemprop
Canonical: https://ross.abutalabs.com/products/chemprop
Homepage: https://chemprop.csail.mit.edu
Language: Python
License: NOASSERTION
License Family: other
Topics: machine-learning, chemistry, neural-networks, drug-discovery
Last push: 2026-08-21T20:08:40+00:00

## Health v2 (maintenance only)
Score: 94/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 98, release rhythm 84, longevity 100
- inputs: {"age_days": 2777, "days_push": 12, "days_rel": 29, "gap_med": 64.5, "n_releases_24m": 11}
- flags: no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 2438, forks 756 (observed 2026-08-28T04:06:51.505561+00:00)

## What it is
Chemprop is a Python package implementing message passing neural networks (MPNNs) for predicting molecular and reaction properties. It provides both a Python API and a command-line interface for training, evaluating, and applying graph neural network models on chemical datasets.

## Use cases
- predict molecular properties from SMILES strings
- train a graph neural network on chemical property data
- predict reaction properties with learned representations
- screen drug candidate molecules for desired properties
- run hyperparameter optimization for molecular property models
- interpret which atoms drive a property prediction

## When to choose
- you need state-of-the-art MPNN models for molecular property prediction
- you want a maintained, well-documented tool with CLI and Python API for cheminformatics ML
- you are doing drug discovery or materials research requiring property prediction

## When to avoid
- you need general-purpose graph neural networks beyond molecular graphs
- you want a no-code GUI tool rather than Python/CLI workflows
- your data is not chemical structures or reactions

## Facets
- artifact type: library
- maturity: active
- function: machine-learning, deep-learning
- domain: chemistry, machine-learning, healthcare
- platform: python, cli, cross-platform
- tags: graph-neural-networks, message-passing-neural-networks, molecular-property-prediction, drug-discovery, cheminformatics

## Member repositories
- chemprop/chemprop (main) score 94

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:06:51.505561+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T02:30:47.231111+00:00, confidence not recorded.
  - readme: https://github.com/chemprop/chemprop (fetched 2026-08-28T04:06:51.505561+00:00, sha 8e1d61de5811)
  - registry_pypi: https://pypi.org/pypi/chemprop/json (fetched 2026-08-29T10:12:47.142384+00:00, sha 0f22af8b67d8)
- Data as of 2026-08-30T08:39:29.467469+00:00.