This paper introduces DiffLinker, an E(3)-equivariant 3D conditional diffusion model designed for molecular linker design. The model aims to address the challenge of connecting disconnected molecular fragments to form chemically relevant candidate drug molecules. Unlike previous methods that can only connect pairs of fragments, DiffLinker can link an arbitrary number of fragments and automatically determines the number of atoms in the linker and their attachment points. The model is evaluated on standard datasets and shown to outperform other methods in terms of synthetic accessibility, drug-likeness, and chemical diversity. Additionally, DiffLinker can be conditioned on target protein pockets, improving the affinity of generated compounds. The method is demonstrated in real-world applications, including fragment-based design of ligands for Hsp90 and IMPDH, and scaffold hopping for improving selectivity for JNKs. The overall goal is to provide a practical tool for molecular linker generation in realistic drug design scenarios.This paper introduces DiffLinker, an E(3)-equivariant 3D conditional diffusion model designed for molecular linker design. The model aims to address the challenge of connecting disconnected molecular fragments to form chemically relevant candidate drug molecules. Unlike previous methods that can only connect pairs of fragments, DiffLinker can link an arbitrary number of fragments and automatically determines the number of atoms in the linker and their attachment points. The model is evaluated on standard datasets and shown to outperform other methods in terms of synthetic accessibility, drug-likeness, and chemical diversity. Additionally, DiffLinker can be conditioned on target protein pockets, improving the affinity of generated compounds. The method is demonstrated in real-world applications, including fragment-based design of ligands for Hsp90 and IMPDH, and scaffold hopping for improving selectivity for JNKs. The overall goal is to provide a practical tool for molecular linker generation in realistic drug design scenarios.