Another big issue is whether g-xTB offers advantages over UMA

We're also going to test it for most of our projects, but we do need stable gradients with continuum solvation for some of the things we do. As far as I know, that's not quite there yet

GitHub - grimme-lab/g-xtb: Development versions of the g-xTB method. Final implementation will not happen here but in tblite (https://github.com/tblite/tblite). Development versions of the g-xTB method. Final implementation will not happen here but in tblite (https://github.com/tblite/tblite). - grimme-lab/g-xtb

Exciting progress on the new extended tight binding method g-xTB by Stefan Grimme, Thomas Froitzheim and colleagues (github.com/grimme-lab/g...)!

g-xTB can be used within ORCA by using the new binaries either via ExtOpt or directly via ORCA's xtb interface.

#ORCA #gxTB #CompChem #QuantumChem

GitHub - grimme-lab/g-xtb: Development versions of the g-xTB method. Final implementation will not happen here but in tblite (https://github.com/tblite/tblite). | Thomas Froitzheim I am very happy to share that we have just released a new and improved version of our g-xTB model — now finally with fully analytic gradients! It took another nine months after the first g-xTB releas...

G-xTB with gradients is now out! #compchem
www.linkedin.com/posts/thomas...

The Syntax of Matter: Synthesis Planning as the Foundation of Generative Chemistry | ChemRxiv Recent advances in deep learning have improved benchmark performance for chemical property prediction, yet reliable transfer to new chemical domains remains limited. A contributing factor is that many models treat molecules primarily as static graphs, ...

The Syntax of Matter: Synthesis Planning as the Foundation of Generative Chemistry | ChemRxiv chemrxiv.org/doi/full/10.... #compchem

Benchmarking physics-inspired machine learning models for transition metal complexes with diverse charge and spin states | ChemRxiv chemrxiv.org/doi/full/10.... #compchem

moTSart: Accelerating Automated Transition State Search with Generative Models in a Low-Data Regime | ChemRxiv #compchem chemrxiv.org/doi/full/10....

Curated digital datasets of acid dissociation constants in dipolar aprotic solvents | ChemRxiv #compchem chemrxiv.org/doi/full/10....

What is the "‡" symbol meaning in a reaction mechanism? I was studying hydroboration from Clayden's Organic Chemistry [1, p. 1281] and the transition state had “‡” symbol in the upper right corner of the activated complex: We know that this is not the...

chemistry.stackexchange.com/questions/13... #compchem

Inverse Design of Inorganic Compounds with Generative AI Machine learning is revolutionizing chemistry. Beyond the value of predictive models accelerating virtual screening, generative AI aims at enabling inverse design, reversing the compound-to-property p...

Inverse Design of Inorganic Compounds with Generative AI
arxiv.org/abs/2604.11827 #compchem

🧪 Can LLMs judge chemical synthesis plans?

Excited to share new work from our industrial postdoc Varvara Voinarovska, in collaboration with Samuel Genheden and Mikhail Kabeshov at AstraZeneca's Molecular AI team!

📄 : doi.org/10.26434/che...

who knew?

oh right we all knew

Link to the paper's not working for me: doi.org/10.26434/che...

Glad to share our new work: RitS (Right into the Saddle): a generative model for direct transition state (TS) prediction.
📄 Paper: chemrxiv.org/doi/full/10.26…
🤠 Code: github.com/isayevlab/RitS
• Flow-matching model with explicit stereochemistry control
• Direct TS generation from SMILEs #compchem

isayevlab (Isayev Lab at CMU) AI4Science and Science4AI

🚨 AIMNet2 MLIP models are now available on @huggingface 😊 Check it out: huggingface.co/isayevlab and updated documentation: isayevlab.github.io/aimnetcentral/… #compchem #compchemsky #chemsky

Avogadro 2.0.0 released Avogadro is a free and open source molecular editor and visualization tool. The new release of Avogadro 2, is a major update, and includes major changes, a revised interface, and...

Avogadro 2.0.0 released

Avogadro is a free and open source molecular editor and visualization tool. The new release of Avogadro 2, is a major update, and includes major changes, a revised interface,...

We Need To Talk About AI... YouTube video by Cool Worlds Podcast

We Need To Talk About AI...
youtu.be/PctlBxRh0p4?...

Vibe physics: The AI grad student Anthropic is an AI safety and research company that's working to build reliable, interpretable, and steerable AI systems.

Vibe physics: The AI grad student
www.anthropic.com/research/vib...

A timely and eloquent essay, well worth reading by anyone in science.

If you've got something really big - 100,000 atoms, a million atoms? I'd really like to try it out.

A Priori Sampling of Transition States with Guided Diffusion Transition states, the first-order saddle points on the potential energy surfaces, govern the kinetics and mechanisms of chemical reactions and conformational changes. Locating them is challenging bec...

A Priori Sampling of Transition States with Guided Diffusion
arxiv.org/abs/2603.25980 #compchem

Release 2026_03_1 (Q1 2026) Release · rdkit/rdkit Release_2026.03.1 (Changes relative to Release_2025.09.1) Acknowledgements (Note: I'm no longer attempting to manually curate names. If you would like to see your contribution acknowledged with you...

We did a new major #RDKit release yesterday. v2026.03.1 is out and the conda-forge builds are available.

As always, especially with major releases, be sure to read the "Backwards incompatible changes" section of the release notes!

github.com/rdkit/rdkit/...

15th RDKit UGM 2026 2026 RDKit User Group Meeting (in person and online)

Free registration for the 2026 #RDKit UGM (both in-person and online attendance) is now open:
www.eventbrite.com/e/1985889262...

Exploring Reaction Condition Space with Expert-Informed Representation Learning | ChemRxiv #compchem chemrxiv.org/doi/full/10....

Density Functional Theory Surrogate Enables Fast and Broad Computational Evaluation of Homogeneous Transition Metal Catalytic Energy Landscapes | ChemRxiv #compchem chemrxiv.org/doi/full/10....

Stochastic tensor contraction for quantum chemistry Jiace Suna and Garnet Kin-Lic Chan (2026) Highlighted by Jan Jensen What this paper lacks in terms of punchy title, it makes up for in conte...

New highlight: Stochastic tensor contraction for quantum chemistry
(Monte Carlo Meets Coupled Cluster: Slashing the Cost of CCSD(T)) www.compchemhighlights.org/2026/03/stoc... #compchem

Machine-Learned Leftmost Hessian Eigenvectors for Robust Transition State Finding The reliable determination of transition states (TSs) benefits from second-order information for robust convergence and validation, but the computational expense of Hessians prohibits their routine us...

Machine-Learned Leftmost Hessian Eigenvectors for Robust Transition State Finding
arxiv.org/abs/2603.21323 #compchem

AI-Predicting Compound Affinity. We Aren't There Yet.

The latest AI/ML tools for co-folding proteins around small molecule ligands can produce impressive results.

Until you start looking at them closely.

Taming T-REX: A Canonical Language for Geometry-Aware Generative Design of Transition-Metal Complexes | ChemRxiv chemrxiv.org/doi/full/10.... #compchem

New preprint: Why some active-learning screens succeed and others fail: identifying active clusters using decision trees doi.org/10.26434/che... #compchem