ZINC-250k

The ZINC-250k dataset contains a collection of 250,000 commercially available compounds used to evaluate molecular generation models for drug design.

Papers using ZINC-250k (13)

Learning-Order Autoregressive Models with Application to Molecular Graph Generation2025 · 20 citesLearning Flexible Forward Trajectories for Masked Molecular Diffusion2025 · 3 citesDriftingMol: Decoder-Coupled Drift for One-Pass Property-Conditional Molecular Generation2026Transport-Coupled Bayesian Flows for Molecular Graph Generation2025 · 1 citesHybridMolGen: a unified framework for goal-directed molecular generation via multi-objective reinforcement learning2026Design and Research of a Dual-Target Drug Molecular Generation Model Based on Reinforcement Learning2026SMolLM: Small Language Models Learn Small Molecular Grammar2026Graph VQ-Transformer (GVT): Fast and Accurate Molecular Generation via High-Fidelity Discrete Latents2025Lift Your Molecules: Molecular Graph Generation in Latent Euclidean Space2024 · 1 citesModular Flows: Differential Molecular Generation2022Will More Expressive Graph Neural Networks do Better on Generative Tasks?2023Classifier-free graph diffusion for molecular property targeting2023AdaMR: Adaptable Molecular Representation for Unified Pre-training Strategy2024