Google Research

Self-supervised foundation model for continuous glucose monitoring, with dual streams separating slow physiological state from transient events.

Open medical multimodal models from Google, built on Gemma 3 with a medically tuned SigLIP vision encoder for clinical text and image understanding.

Medically tuned SigLIP encoder from Google that maps medical images and text into one embedding space for zero-shot classification and retrieval.

Sensor-language foundation models aligning wearable biosignals with text for zero-shot activity recognition, retrieval, and sensor captioning.

Single-cell foundation model reading scRNA-seq profiles as ranked gene-name sentences, scaled on Gemma-2 for annotation, reasoning and drug screens.

Open therapeutics foundation models from Google, built on Gemma-2, for drug-discovery property prediction and conversational reasoning.

Computed tomography embedding model that compresses a whole DICOM CT volume into a 1,408-number vector for data-efficient downstream classifiers.

Wearable sensor foundation model pretrained on heart rate, accelerometer, skin temperature and other channels for activity recognition and imputation.

Chest X-ray embedding model built on ELIXR, producing image and image-text embeddings for data-efficient and zero-shot radiograph classification.

Family of medical multimodal models built on Gemini, adding uncertainty-guided web search, custom modality encoders, and long-context EHR reasoning.

Google's dermatology image embedding model that produces 6144-dimensional embeddings for data-efficient skin-condition classifiers.

Histopathology foundation model that encodes 224x224 H&E patches into compact 384-dimensional embeddings for tumor and biomarker classifiers.

Health acoustics foundation model that turns short clips of coughs and breaths into embeddings for building acoustic biomarker models with less data.

Google's generalist multimodal biomedical AI that encodes clinical text, medical images, and genomics with a single set of weights across 14 tasks.

Sparse attention transformer that extends BERT to 8x longer sequences via random, local, and global attention, with genomic sequence applications.