AllChem generates ~5 million synthons by recursively applying ~100 reactions to ~7,000 building blocks, combinatorially representing up to 10^20 complete structures with an A-B-C topology. Topomer shape similarity enables efficient searching of this space, and every hit comes with a proposed synthetic route.
CHX8 exhaustively enumerates all mathematically feasible hydrocarbons with up to eight carbon atoms (77,524 structures), then DFT-optimizes them to identify 31,497 stable molecules. A universal relative strain energy (RSE) metric referenced to cyclohexane serves as a synthesizability proxy. CHX8 covers 16x more C8 hydrocarbons than GDB-13 and reveals that over 90% of novel structures should be synthetically accessible.
FDB-17 contains 10 million fragment-like molecules selected from GDB-17’s 166.4 billion entries. Fragment-likeness filters reduce GDB-17 by 36x to 4.6 billion molecules, then even sampling across (HAC, heteroatoms, stereocenters) triplets produces a 460x further reduction to a manageable, diverse library enriched in 3D-shaped molecules.
GDBMedChem applies medicinal chemistry-inspired functional group and structural complexity filters to GDB-17, reducing 166.4 billion molecules to 17.8 billion, then evenly samples across molecular size, stereochemistry, and polarity to produce 10 million drug-like molecules. 97% of its substructures are absent from known molecule databases.
QM9 provides B3LYP/6-31G(2df,p)-level geometric, energetic, electronic, and thermodynamic properties for 133,885 small organic molecules (up to 9 heavy atoms of C, N, O, F) drawn from the GDB-17 chemical universe. It is one of the most widely used benchmarks in molecular machine learning.
VEHICLe (Virtual Exploratory Heterocyclic Library) is a complete enumeration of 24,867 mono- and bicyclic heteroaromatic ring systems built from C, N, O, S, and H. Of these, only 1,701 have ever appeared in published compounds. A random forest classifier trained on known vs. unknown ring systems predicts that over 3,000 additional ring systems are synthetically tractable.
VQM24 exhaustively enumerates all neutral closed-shell molecules with up to 5 heavy atoms from C, N, O, F, Si, P, S, Cl, Br, yielding 258k constitutional isomers and 578k conformers (836k total). Properties are computed at the wB97X-D3/cc-pVDZ level, with diffusion QMC energies for 10,793 molecules up to 4 heavy atoms. ML models show up to 8x higher errors than on QM9, making VQM24 a more challenging benchmark.
Shen et al. empirically analyze how different domain combinations and deduplication strategies in the SlimPajama dataset affect 1.3B model performance. Global deduplication across sources outperforms local deduplication, and increasing domain diversity consistently improves average accuracy, with findings transferring to 7B scale.
ChemLLM presents a comprehensive framework for chemistry-specific language modeling, including a 7M-sample instruction tuning dataset (ChemData), a 4,100-question benchmark (ChemBench), and a two-stage fine-tuned model that matches GPT-4 on core chemical tasks.
ChemBench introduces an automated benchmark of 2,700+ chemistry questions to evaluate LLMs against human expert chemists, revealing that frontier models outperform domain experts on average while struggling with basic tasks and confidence calibration.
DOCKSTRING bundles an AutoDock Vina wrapper, a 260K-molecule docking dataset across 58 protein targets, and pharmaceutically relevant benchmarks for regression, virtual screening, and de novo design.
MoleculeNet introduces a large-scale benchmark suite for molecular machine learning, curating over 700,000 compounds across 17 datasets with standardized metrics, data splits, and featurization methods integrated into the DeepChem open-source library.