On the personal side, after participating for a couple of years as an attendee & speaker, I am now joining the amazing organising team 🤩 with Anoop K, @martoskreto, Emily Jin, @KevinJablonka, Stefano Martiniani, Rocio Mercado, @santiagomiret . Check out the website for more.

Fellow AI for Materials researchers, . The @NeurIPSConf #AI4Mat workshop's deadline is in ~20 days ! It's from experience a fantastic opportunity to submit your ongoing work, get meaningful reviews and discuss it with the community ! Whether you do AI-guided materials design,.

🤗⚛️ Join the @entalpic_ai Team ⚛️🤗 . At Entalpic, we combine MLIP, generative models, quantum simulations and lab experiments to optimise carbon intensive industrial processes and develop the next-generation materials. ⚠️ We are specially looking for ⚠️ .* (Senior and Intern)

RT @martoskreto: AI4Mat is back for NeurIPS! time to crystallize those ideas and make a solid-state submission by august 22, 2025 💪. new th….

Thank you so much #AI4X and Singapour ! . A week of interesting discussions with the AI for Material/Chemistry community, experimental lab visits, amazing talks, client meetings, fancy rooftops, and more. Concluded by our presentation of Entalpic’s AI materials discovery

RT @jrib_: Matbench Discovery is out in Nature Machine Intelligence @. Paper: Leaderboard: .

RT @vdbergrianne: 🚀 After two+ years of intense research, we’re thrilled to introduce Skala — a scalable deep learning density functional t….

I love seeing the ongoing discussions on equivariance & energy conservation for ML Interatomic Potentials. Whether to enforce it explicitly or not, exploring the trade-off, understanding what's needed for a specific problem. Would recommend reading the thoughts of @chaitjo,.

RT @bwood_m: We released the Open Molecules 2025 (OMol25) Dataset last week! 🚀🧪 OMol25 is a large (100M+) and diverse molecular DFT dataset….

We didn't have the pleasure to attend @iclr_conf this year, to present our LeMat-Bulk paper - accepted in the AI4Mat workshop. But it's available here --> Please let us know if you have any feedback, the full version still under review in a journal🤞.

5/ 🎯 Want to help shape the next releases?. Join our Slack & working groups:.• Generative models for materials.• LLMs for synthesis extraction.• Multi-material datasets.• …or pitch your idea!. Let’s build the foundation models, benchmarks, and datasets that fuel tomorrow’s.

4/ 🙌 Made possible by a joint effort from:.@huggingface @Thom_Wolf 🤗 + contributors Ali Ramlaoui, Inel Djafar, Martin Siron, Amandine Rossello, Alexandre Duval, and the growing #LeMaterial community. Without forgetting the amazing initial effort from Materials Project,.

3/ 📊 And… a benchmark for material structure hashing!. You can use it to generate unique fingerprints for crystal structures and easily compare and de-duplicate data. Add your hashing/similarity method and benchmark against others!.→

2/ 🛠️ We are also releasing the code that:.• Fetches and updates the datasets regularly.• Cleans & standardizes data.• Makes it easy for the community to add more datasets. →

1/ ⚛️ LeMaterial goes to Trajectories ⚛️. After releasing LeMat-Bulk - a crystals database of 6.7M structures - we’re excited to launch LeMat-Traj, a collection of ~120M structures from 3.7M atomic relaxation trajectories, compiled from Materials Project, Alexandria and OQMD.

The more time I spend in experimental materials labs, the more impressed I am by what we have accomplished as a civilisation — designing complex chemical processes with incomplete knowledge of underlying reaction mechanisms. The associated complexity is quite humbling tbh, but I.

Hi together, . We are having our second LeMaterial community meeting tomorrow (April 10) at 6PM CET | 9AM LA 📅. We will be discussing code / paper / datasets releases + ongoing working group progress (generative models, LLMs, data) 🤩. 👉 To join:

🚨 Job alert 🚨. @entalpic_ai is hiring a Computational Chemist to lead ML-driven material discovery for liquid-phase chemical systems -- focusing on redox-active molecules, electrolytes, and molecule-solvent interactions. You will work at the interface of ML, computational.

Thanks SmartDigicat for catalysing discussions with chemists and physicists about the challenges of digitalised chemistry. Some examples: . - cost and reliability of lab equipments .- standardising experimental & computational databases (check out effort)

Wow ! Zurich was intense and buzzing with AI for materials discovery ! In 1 day, we .🔬Visited SwissCat+ high-throughput facilities for heterogenous catalysis .🎤Presented @entalpic_ai research, and had a super insightful discussions with ETHZ students & Jonathan Schmidt