Edge Devices
Deploy trained BCI models to mini edge hardware. Stream live biosignals, run on-device inference, and build real-world neurotechnology applications — we're actively working on this.
We are actively working on support for devices like Raspberry Pi and NVIDIA Jetson Nano. Nothing is connected yet — but it's coming. Stay tuned.
What's coming
Deploy Trained Models to Edge
Export any pipeline you train in BCILattice as an optimized runtime and push it directly to a connected edge device.
Live Signal Acquisition on Device
Acquire EEG, EMG, or fNIRS from a device-connected amplifier, stream to BCILattice over the local network, or classify fully on-device.
On-Device Real-Time Inference
Run classification entirely on the edge device with lightweight ONNX runtimes or GPU-accelerated inference where available.
ONNX & TFLite Export
One-click export to ONNX or TFLite for maximum compatibility. Works with any model trained in BCILattice, no manual conversion needed.
Local Network Streaming
Stream live biosignal data from an edge device to your BCILattice workstation over a local network using LSL or a lightweight custom protocol.
Air-Gapped Compatible
Deploy to edge hardware in completely offline environments. No internet connection required at any point, from training to deployment.
ONNX and TFLite export from BCILattice are already working. Device connectivity, on-device runtime, and LSL streaming over local network are under active design.
Get notified when it ships
Sign up to get early access and be the first to deploy BCI models to edge hardware.