Introducing SynthID Text

SynthID Detector — a new portal to help identify AI-generated content

Learn about the new SynthID Detector portal we announced at I/O to help people understand how the content they see online was generated.

Strategy-first synthesis planning for complex natural products

arXiv:2608. 07454v1 Announce Type: cross Abstract: The total synthesis of a complex molecule is among the most demanding intellectual and experimental feats in chemistry: a chemist must plan many steps ahead for how to assemble simple building blocks into an intricate target, devise backup strategies, and anticipate procedural challenges.

How to generate text: using different decoding methods for language generation with Transformers

Learning from flowsheets: A generative transformer model for autocompletion of flowsheets

arXiv:2208. 00859v2 Announce Type: replace Abstract: We propose a novel method enabling autocompletion of chemical flowsheets.

LeMat-Synth: a multi-modal toolbox to curate broad synthesis procedure databases from scientific literature

arXiv:2510.26824v2 Announce Type: replace-cross Abstract: Wide access to advanced experimental methods in materials science has given rise to an abundance of procedural knowledge, which is scattered...

Navigating the challenges and opportunities of synthetic voices

We’re sharing lessons from a small scale preview of Voice Engine, a model for creating custom voices.

Assisted Generation: a new direction toward low-latency text generation

Introducing Codex

Improving synthesis prediction of small molecules at scale with RetroChimera

The article discusses RetroChimera, a predictive model developed by Microsoft Research that improves the prediction of small molecule synthesis at scale. It highlights how the model can accelerate chemical synthesis, enabling researchers to explore a broader range of molecules. The post appears in a Nature paper and emphasizes the model’s potential to streamline the creation of custom-made molecules for medicine, materials, and agriculture.

Introducing the Synthetic Data Generator - Build Datasets with Natural Language

MolGVR: A Chemistry-Grounded Framework for Text-to-Molecule Generation

arXiv:2607. 29479v1 Announce Type: new Abstract: Text-to-molecule generation is typically formulated as a one-shot sequence generation problem, where a model directly maps target descriptions to molecular representations.

When Single Answer Is Not Enough: Rethinking Single-Step Retrosynthesis Benchmarks for LLMs

arXiv:2602. 03554v2 Announce Type: replace-cross Abstract: Recent progress has expanded the use of large language models (LLMs) in drug discovery, including synthesis planning.