Google DeepMind Unveils IsoDDE Drug Discovery Engine, Updates Gemini Deep Think

Isomorphic Labs, a subsidiary of DeepMind, introduced the IsoDDE engine for drug development. In complex tests, the innovation outperformed AlphaFold 3 in prediction accuracy by a factor of two.

AlphaFold 3 was a major breakthrough as it could predict the three-dimensional structures of proteins and their interactions with molecules. IsoDDE demonstrates an entirely new level: the model predicts binding strength (affinity) more accurately than traditional methods, the engine can identify hidden structures ("pockets") in proteins where drugs can bind, and it supports a wide range of complex molecules, including antibodies and large biological structures. The company stated that IsoDDE offers a scalable framework for AI-driven drug design, providing the prediction accuracy necessary to work with new biological systems with unprecedented reliability.

Isomorphic Labs is re-imagining what it takes to bring life-saving medicines faster with AI. The company now has 17 active programmes and early-stage cancer therapies in pre-clinical trials. Drug development typically takes on average 10 years with only a 10% success rate. The approach involves doing search and hypothesis searching in silico, which is hundreds to thousands of times more efficient than doing it in a wet lab.

Google has updated the reasoning mode of Gemini 3 Deep Think, positioned as a solution for complex tasks in science and engineering. In tests, the model outperformed OpenAI's GPT-5.2 and Anthropic's Claude Opus 4.6, including in ARC-AGI-2 with visual puzzles, MMMU-Pro for evaluating multimodal capabilities, Elo 3455, and the "Last Exam of Humanity." The company stated that Gemini 3 Deep Think was updated in close collaboration with scientists and researchers to tackle complex scientific challenges—where tasks often lack clear boundaries or a single correct solution, and data is incomplete.

Gemini 3 Deep Think demonstrates advanced results in mathematics and programming, and performs excellently in natural sciences, including chemistry and physics. The updated mode solves problems at the level of gold medalists in international olympiads. In the CMT-Benchmark, the model scored 50.5%, confirming deep knowledge in theoretical physics. Beyond advanced metrics, Deep Think is geared towards practical application: it helps researchers interpret complex data and engineers model physical systems through code. The new Deep Think is available in the Gemini app for Google AI Ultra subscribers and Gemini API for select developers.

DeepMind introduced the AI agent Aletheia, which set a new record in the IMO-ProofBench Advanced benchmark, solving 91.9% of tasks. The test is considered one of the most challenging in mathematics. The neural network is built on the Gemini Deep Think platform and is equipped with a verification module: it identifies errors in draft solutions and initiates an iterative process of refinement. A key feature of the agent is its ability to recognize the impossibility of solving a problem, significantly saving researchers' time.

Aletheia uses Google Search to navigate complex scientific materials, preventing the use of false links and computational errors when working with scientific materials. Among the model's achievements: complete generation of a scientific paper calculating structural constants in arithmetic geometry, collaborative proof of estimates for systems of interacting particles (independent sets) with a human, and autonomous solution of four problems from the Erdős list, one of which was previously considered open. DeepMind emphasized that Aletheia's success confirms the relevance of scaling laws: in proof-based mathematics, quality continues to improve through the effective application of agents.

AlphaFold addressed a 50-year grand challenge in biology—predicting protein structures from amino acid sequences. The breakthrough compressed decades of research into a database that is now open to the world. Today, over 3 million researchers in more than 190 countries are using it to develop malaria vaccines, fight antibiotic resistance and much more. In July 2022, the AlphaFold algorithm predicted the structures of over 200 million proteins, encompassing nearly all known compounds found in plants, bacteria, and animals.

Google is establishing a full-stack AI hub in Visakhapatnam, India, part of a $15 billion infrastructure investment in India. When finished, this hub will house gigawatt-scale compute and a new international subsea cable gateway, bringing jobs and the benefits of cutting-edge AI to people and businesses across India. Google is also building a vast network of subsea fiber optic cables, including four new systems between the U.S. and India, as part of the America-India Connect Initiative.

In 2023, Google unified DeepMind and Google Brain to consolidate talent and compute. Google DeepMind functions as the engine room of Google, powering the company's broader AI capabilities. The company has trained 100 million people in digital skills, and a new Google AI Professional Certificate will help people master AI in their jobs, available globally.