Nothing you type, record, or upload here is sent to Tathastha Labs or any server.
A named-entity model finds people, organizations, and locations in raw clinical text — patient and clinician names, facility names — the same category of model that underlies de-identification and structured-extraction pipelines.
Model: Xenova/bert-base-NER · ~45 MB first run
This is a general-purpose vision classifier, not a diagnostic model — we're showing the mechanics of an in-browser imaging pipeline (upload → preprocess → inference → ranked output), the same shape as the modality-specific models we build for imaging engagements.
Model: Xenova/vit-base-patch16-224 · ~90 MB first run
A biomedical NER model from OpenMed's open-source model family, fine-tuned specifically to recognize disease and condition mentions in clinical text — the kind of model that feeds structured problem lists and registries.
Model: OpenMed-NER-DiseaseDetect-TinyMed-135M · ~65 MB first run
Speech is transcribed with Whisper, then a small instruct model drafts a structured clinical summary from the transcript — the same two-stage pattern behind our ambient scribe work, running entirely on-device here. This is a 0.5B-parameter model for browser-friendly size, so treat the formatting as a rough draft, not a polished SOAP note — production engagements use larger, evaluated models for that.
Models: Xenova/whisper-tiny.en (~40 MB) +
onnx-community/Qwen2.5-0.5B-Instruct (~300 MB) · first run only