> ## Documentation Index
> Fetch the complete documentation index at: https://docs.nimbusbci.com/llms.txt
> Use this file to discover all available pages before exploring further.
# Nimbus BCI Engine
> Production-ready Bayesian BCI inference in Python and Julia with real-time streaming, uncertainty quantification, and models for motor imagery, P300, and SSVEP.
*Production-ready Bayesian BCI inference for Python and Julia.*
Nimbus provides fast probabilistic inference for brain-computer interface applications. Use it for motor imagery, P300, SSVEP, adaptive BCI, and research workflows that need confidence-aware decisions.
## Start Here
Local sklearn-compatible workflows with `nimbus-bci`.
RxInfer-backed workflows with `NimbusSDK.jl`.
Choose between LDA, QDA, Softmax, Probit, and STS.
## Why Nimbus
Compare classical LDA, SVM, and deep learning with Nimbus — including \~10× faster online updates.
Predictions include posterior probabilities and confidence scores.
Batch and streaming workflows for low-latency BCI systems.
Bayesian model families tuned for neural feature data.
Validation, quality gates, diagnostics, and deployment patterns.
## Models
| Model | SDK | Best For |
| --------------- | -------------- | ----------------------------------------------------------- |
| `NimbusLDA` | Python + Julia | Fast baseline for well-separated CSP or bandpower features. |
| `NimbusQDA` | Python + Julia | Overlapping classes and class-specific covariance. |
| `NimbusSoftmax` | Python | Multiclass nonlinear boundaries with optional JAX install. |
| `NimbusProbit` | Julia | Julia-native Bayesian multinomial probit workflows. |
| `NimbusSTS` | Python | Non-stationary sessions and latent-state adaptation. |
## Core Workflow
```text theme={null}
EEG acquisition -> preprocessing -> Nimbus model -> confidence-aware action
```
Nimbus expects preprocessed features rather than raw EEG. Start with [Preprocessing Requirements](/inference-configuration/preprocessing-requirements) if you are setting up CSP, bandpower, ERP, or external tool exports.
## Documentation Map
Choose Python or Julia and run first inference.
Preprocessing, normalization, batch, streaming, and error handling.
Compact recipes and higher-level application patterns.
SDK-specific functions, classes, and data structures.
## Common Questions
Python does not require an API key. Julia requires an API key to install and use the commercial core.
No. Nimbus expects features produced by preprocessing pipelines such as CSP, bandpower, or ERP extraction.
Use Python for sklearn/MNE workflows and local development. Use Julia for RxInfer-backed workflows and Julia-native model tooling.
## Next Read