Mol3D

TabLab-Hub

See a molecule in 2D and 3D0.33.0

Random

What Mol3D is

Mol3D shows a molecule in 2D and 3D.
Type or paste SMILES, open a file, or draw it.
Build 3D makes a 3D structure; Optimize minimizes it; drag atoms or turn bonds by hand.
It runs in your browser tab; nothing is uploaded elsewhere.

What should I not conclude?

A generated conformer is not an experimental structure. A force-field energy is not a binding affinity: it compares conformers of the same molecule only. Charged and metal-containing species are not validated beyond what is tested.

Is anything uploaded?

No. The page sends anonymous counts to its own address (one when opened, one per structure shown, one per save): a request with no data; the server keeps site, day, kind and country. The editor opens no other site.

Which inputs work?

SMILES, one per line with an optional name; or SDF, MOL, XYZ or PDB text, pasted or opened. Build 3D and Optimize need bonds: not XYZ or PDB. Over 5 MB or 100 molecules is refused.

How do I open a molecule?

Open a file, draw it in Draw, or paste it in the text box and press Show. Its SMILES appears under the 2D view.

How much loads?

Up to 27.8 MB on a first visit (22.1 MB without OpenFF Sage 2.3.0), kept for the next: the page 1.1 MB, the 2D engine 7.6 MB, the 3D engine 4.9 MB, OpenFF Sage 2.3.0 5.7 MB; when needed the editor 2.5 MB, a Random list 1.5 MB (ChEMBL) or 4.5 MB (PDB).

How do I cite it?

Cite the web address: https://mol3d.com/

Credits

Data
your own structures stay yours · Random: a sample of ChEMBL 37 (www.ebi.ac.uk/chembl, CC BY-SA 3.0, share-alike) and of the PDB's ligands (wwPDB and RCSB PDB, CC0 1.0) · OpenFF Sage 2.3.0 parameters (Open Force Field Initiative, CC BY 4.0; changed: three sulfonyl angle terms, four alkane torsion terms, three acid-hydrogen van der Waals terms, five N-H torsion terms (amide, urea, thiourea)) and OpenFF NAGL partial charges (openff-gnn-am1bcc-1.0.0, checkpoint sha256 7981e7f5b0b1e424, Open Forcefield Group, CC BY 4.0): a network trained on AM1-BCC charges, so the charges here are its estimates, not AM1-BCC calculations; the model's own lookup table of reference charges is not shipped; implicit water: the GB-OBC model with its published constants (Onufriev, Bashford and Case 2004; Hawkins, Cramer and Truhlar 1996; Schaefer and Karplus 1996; radii after Bondi 1964), written here from the papers, none of OpenMM's code or files used; the water is not fitted with Sage's parameters or the charge model
Code
RDKit BSD-3-Clause · OpenChemLib BSD-3-Clause · JSME BSD-3-Clause (modified: outgoing links removed) · 3Dmol.js BSD-3-Clause · licences
Cite
Landrum (RDKit) · Sander et al. 2015 (OpenChemLib) · Halgren 1996 and 1999 (MMFF94, MMFF94s) · Boothroyd et al. 2023 (OpenFF Sage) · Onufriev et al. 2004 (GB-OBC water, mbondi2 radii) · Hawkins et al. 1996 (descreening) · Schaefer and Karplus 1996 (surface term) · Bienfait and Ertl 2013 (JSME) · Rego and Koes 2015 (3Dmol.js) · Zdrazil et al. 2024 (ChEMBL) · Burley et al. 2023 (RCSB PDB)