Nimbus BCI Engine

Production-ready Bayesian BCI inference with NimbusLDA and NimbusQDA, plus NimbusSoftmax / NimbusSTS (Python) and NimbusProbit (Julia) Nimbus provides fast, probabilistic inference for Brain-Computer Interface applications in Python and Julia. The SDKs deliver sub-30ms inference latency with full uncertainty quantification through Bayesian LDA, Bayesian QDA, Bayesian Multinomial, and Bayesian STS implementations.

Choose your SDK:

Python SDK (nimbus-bci)

sklearn-compatible classifiers with MNE-Python integration

Julia SDK (NimbusSDK.jl)

Built on RxInfer.jl for maximum performance

Start Here

Python Quickstart

Start with local sklearn-compatible workflows in nimbus-bci.

Julia Quickstart

Start with NimbusSDK.jl and RxInfer-powered inference.

Model Selection

Choose the right Bayesian model for your BCI task.

Why Nimbus BCI

Brain-Computer Interfaces face unique challenges: noisy neural signals, inherent uncertainty, real-time requirements, and the need for explainability in medical applications. Current BCI Challenges:

High Latency: Standard processing takes 200ms+ for trial classification
No Uncertainty Quantification: Deterministic outputs without confidence measures
Limited Adaptability: Cannot handle changing brain states or signal quality
Black Box Models: Deep learning lacks explainability for FDA approval

Nimbus Solution:

Fast Inference: 10-20ms per trial with Bayesian models
Uncertainty Quantification: Full posterior distributions, not just point estimates
Training & Calibration: Subject-specific personalization in minutes
Explainable: White-box probabilistic models for medical compliance
Production-Ready: Batch and streaming modes, quality assessment, performance tracking

Python SDK Quickstart

Get started with nimbus-bci in Python

Julia SDK Quickstart

Get started with NimbusSDK.jl in Julia

Core Features

Fast Bayesian Inference

NimbusLDA and NimbusQDA models with 10-20ms inference latency using RxInfer.jl reactive message passing

Training & Calibration

Train custom models on your data or calibrate pre-trained models with 10-20 trials per subject

Real-time Streaming

Process EEG chunks as they arrive with chunk-by-chunk inference and weighted aggregation

Uncertainty Quantification

Full posterior distributions with confidence scores for quality assessment and rejection

Implemented Models

Nimbus provides production-ready Bayesian models. Some are available in both SDKs, and some are SDK-specific:

Bayesian LDA (NimbusLDA)

Bayesian Linear Discriminant Analysis with shared covariance. Fastest (10-15ms), ideal for well-separated classes like motor imagery. Python: NimbusLDA, Julia: NimbusLDA.

Bayesian QDA (NimbusQDA)

Bayesian Quadratic Discriminant Analysis with class-specific covariances. More flexible (15-25ms), better for overlapping distributions like P300. Python: NimbusQDA, Julia: NimbusQDA.

Bayesian Multinomial (NimbusSoftmax / NimbusProbit)

Non-Gaussian multinomial classifier family. Python: NimbusSoftmax. Julia: NimbusProbit.

Bayesian STS (NimbusSTS)

State-space model with EKF-style inference for drift/non-stationarity and online learning. Python: NimbusSTS.

BCI Paradigms Supported

Motor Imagery

2-4 class classification

Left/right hand
Hands/feet/tongue
70-90% accuracy
NimbusLDA recommended

P300

Binary classification

Target/non-target
Speller applications
80-95% accuracy
NimbusQDA recommended

SSVEP

Multi-class frequency

2-6 target frequencies
High accuracy (85-98%)
Works with both models

Quick Example

Python
Julia

from nimbus_bci import NimbusLDA
import numpy as np

# 1. Create and fit classifier
clf = NimbusLDA()
clf.fit(X_train, y_train)

# 2. Predict with uncertainty
predictions = clf.predict(X_test)
probabilities = clf.predict_proba(X_test)

# 3. Works with sklearn pipelines
from sklearn.pipeline import make_pipeline
from sklearn.preprocessing import StandardScaler

pipe = make_pipeline(StandardScaler(), NimbusLDA())
pipe.fit(X_train, y_train)

# 4. Online learning
clf.partial_fit(X_new, y_new)

using NimbusSDK

# 1. One-time setup: Install core with your API key
NimbusSDK.install_core("nbci_live_your_key")

# 2. Load pre-trained model
model = load_model(NimbusLDA, "motor_imagery_4class_v1")

# 3. Prepare preprocessed features (CSP, bandpower, etc.)
data = BCIData(csp_features, metadata, labels)

# 4. Run inference
results = predict_batch(model, data)

# 5. Analyze
println("Accuracy: $(sum(results.predictions .== labels) / length(labels))")
println("Mean confidence: $(mean(results.confidences))")
println("ITR: $(calculate_ITR(accuracy, 4, 4.0)) bits/minute")

Use Cases

Assistive Technologies

Wheelchair control, prosthetic limbs, communication devices with explainable, FDA-ready probabilistic models

Research Platforms

Academic research with training capabilities, subject-specific calibration, and comprehensive performance metrics

Neurofeedback

Real-time brain state monitoring with streaming inference and confidence-based quality control

Gaming & Wellness

Low-latency brain control for immersive experiences with adaptive difficulty based on confidence

SDK Architecture

EEG Hardware → Preprocessing → Nimbus SDK → Application
   (LSL/BrainFlow)  (MNE/EEGLAB)  (Python/Julia)  (Your Code)

Key Design:

Local Inference: All processing on your machine (Python SDK is fully local)
Privacy: Your EEG data never leaves your computer
Speed: No network latency, consistent <20ms performance
Offline Capable: Python SDK works completely offline

Julia SDK API Role:

Authentication and licensing
Pre-trained model distribution
Optional analytics logging

Getting Started

Python
Julia

Install Python SDK

pip install nimbus-bci

Preprocess EEG

Use MNE-Python to extract features (CSP, bandpower, ERP)

Train & Predict

Use sklearn-compatible API to train models and make predictions

Deploy

Integrate with sklearn pipelines, streaming inference, or real-time applications

Install Julia SDK

using Pkg
Pkg.add("NimbusSDK")

# One-time setup: Install core with your API key
using NimbusSDK
NimbusSDK.install_core("your-api-key")

Get API Key

Contact hello@nimbusbci.com for your API key and license tier

Preprocess EEG

Use MNE-Python, EEGLAB, or your tool to extract features (CSP, bandpower, ERP)

Run Inference

Load model, process features, get predictions with confidence scores

Documentation

Python SDK

sklearn-compatible Bayesian classifiers

Julia SDK

RxInfer.jl-based inference engine

Model Training

Train custom models on your data

Preprocessing Guide

Required EEG preprocessing steps

Code Examples

Working Python and Julia examples

API Reference

Authentication and model registry

Next Read

Python SDK

Understand local Python workflows and integration patterns.

Julia SDK

Explore Julia API workflows and model operations.

Cloud API

Review authentication, licensing, and endpoint architecture.

Common Questions

Which SDK should I choose: Python or Julia?

Use the Python SDK Quickstart if you want sklearn-compatible workflows, MNE integration, and fully local development with pip.Use the Julia SDK Quickstart if you want RxInfer.jl-native performance, model registry access, and Julia-first pipelines.

Which Nimbus model should I start with?

Start with Bayesian LDA for most motor-imagery pipelines and fast baseline performance.Move to Bayesian QDA for overlapping class distributions (for example, P300), and use Bayesian STS for non-stationary sessions that drift over time.

Can Nimbus process raw EEG directly?

Nimbus expects preprocessed features rather than raw EEG. See Preprocessing Requirements and Feature Normalization for the recommended pipeline.

Performance

Metric	NimbusLDA	NimbusQDA	NimbusSoftmax (Py)	NimbusSTS (Py)
Inference Latency	10-15ms	15-25ms	15-25ms	20-30ms
Training Time	10-30s	15-40s	15-40s	15-40s
Calibration Time	5-15s	8-20s	8-20s	8-20s
Memory Usage	Low	Moderate	Moderate	Moderate
Accuracy	70-90% (MI)	80-95% (P300)	70-85% (Complex)	75-90% (Adaptive)

All measurements on standard CPU (no GPU required).

Technology Stack

Python SDK:

Core: Python 3.10+, NumPy, JAX, NumPyro
Integration: scikit-learn pipelines, MNE-Python
Models: NimbusLDA, NimbusQDA, NimbusSoftmax, NimbusSTS

Julia SDK:

Core: Julia 1.9+
Inference Engine: RxInfer.jl - Reactive message passing
Models: NimbusLDA, NimbusQDA, NimbusProbit

Common:

Preprocessing: MNE-Python, EEGLAB, BrainFlow
API: TypeScript/Vercel serverless (Julia SDK authentication)

What’s Included

Python SDK (nimbus-bci): ✅ sklearn-compatible classifiers: NimbusLDA, NimbusQDA, NimbusSoftmax, NimbusSTS ✅ MNE-Python integration for EEG preprocessing ✅ Streaming inference for real-time BCI ✅ Online learning with partial_fit() ✅ Comprehensive metrics and diagnostics Julia SDK (NimbusSDK.jl): ✅ RxInfer.jl-based models: NimbusLDA, NimbusQDA, NimbusProbit ✅ Pre-trained model distribution ✅ Batch and streaming inference ✅ Training and calibration ✅ Quality assessment and ITR calculation

Support

Email Support

hello@nimbusbci.com - Technical support and inquiries

Book a Demo

See Nimbus in action with live demonstration

API Status

Check API availability and version

GitHub

NimbusSDK.jl source code, issues, and examples

License

Python SDK (nimbus-bci): Proprietary license with free evaluation and academic tiers. See Python SDK Installation for details. Julia SDK (NimbusSDK.jl): Commercial software with tiered licensing:

Free: 10K monthly inferences, basic models
Research: 50K monthly inferences, all features
Commercial: 500K monthly inferences, priority support
Enterprise: Unlimited, custom models, on-premise deployment

Contact hello@nimbusbci.com for licensing information.

Built with ❤️ for the neurotechnology community Nimbus BCI Engine - Bringing Bayesian inference to Brain-Computer Interfaces

Getting Started

Core Concepts

Models & Algorithms

Configuration & Setup

Examples

Development

Python SDK

Julia SDK

Python SDK (nimbus-bci)

Julia SDK (NimbusSDK.jl)

​Start Here

Python Quickstart

Julia Quickstart

Model Selection

​Why Nimbus BCI

Python SDK Quickstart

Julia SDK Quickstart

​Core Features

Fast Bayesian Inference

Training & Calibration

Real-time Streaming

Uncertainty Quantification

​Implemented Models

Bayesian LDA (NimbusLDA)

Bayesian QDA (NimbusQDA)

Bayesian Multinomial (NimbusSoftmax / NimbusProbit)

Bayesian STS (NimbusSTS)

​BCI Paradigms Supported

Motor Imagery

P300

SSVEP

​Quick Example

​Use Cases

Assistive Technologies

Research Platforms

Neurofeedback

Gaming & Wellness

​SDK Architecture

​Getting Started

​Documentation

Python SDK

Julia SDK

Model Training

Preprocessing Guide

Code Examples

API Reference

​Next Read

Python SDK

Julia SDK

Cloud API

​Common Questions

​Performance

​Technology Stack

​What’s Included

​Support

Email Support

Book a Demo

API Status

GitHub

​License

Start Here

Why Nimbus BCI

Core Features

Implemented Models

BCI Paradigms Supported

Quick Example

Use Cases

SDK Architecture

Getting Started

Documentation

Next Read

Common Questions

Performance

Technology Stack

What’s Included

Support

License