ChatPPG.
Join free

Model family 02 of 04

Multimodal

Multimodal models learn from several sensor streams together, for example PPG with ECG, motion or respiration, and map them into a shared representation. Training modalities can differ from what the model needs at inference, so check each record for its supported inputs.

Signals → shared representation
Models
22
Open weights
5
Benchmarks
9
Years
24–26
  • PPG · 18
  • ECG · 13
  • IMU · 10
  • EEG · 6
  • Resp · 4
  • Temp · 4
02
Conceptual diagram of the family's data flow, not a model architecture.

22 multimodal models

Newest first · one primary role per record

Where these models are benchmarked

All leaderboards

Datasets these models use

All datasets

Families are editorial primary-role groupings reviewed against each model record, not an architecture taxonomy. A model appears in one family even when it could fit more than one. Diagrams are conceptual.

Other families

All 162 models

Flow previews are paused. Each diagram shows the complete process.

0143 modelsSignal encodersEncode a waveform into features for downstream tasks.Explore models0222 modelsMultimodalLearn relationships across pulse, heart and motion signals.You are here0316 modelsSensor–languageAlign recordings with text for retrieval or interpretation.Explore models042 modelsForecastingUse past measurements to estimate what may come next.Explore models

Conceptual diagrams, not model architectures. One primary role per record; families can overlap.

Other / unclassified 79