A
securityA
licenseA
qualityA Model Context Protocol server that provides basic mathematical and statistical functions to LLMs, enabling them to perform accurate numerical calculations through a simple API.
Last updated -
13
5
52
MIT License
Uses FastAPI to expose the Bayesian MCP server API endpoints for model creation, belief updating, prediction generation, model comparison, and visualization.
Provides installation instructions for cloning the repository from GitHub to set up the Bayesian MCP server.
Utilizes NumPy for numerical computations in Bayesian inference, handling arrays and mathematical operations for probabilistic models.
Built on Python 3.9+ and leverages Python's ecosystem for implementing Bayesian reasoning and inference capabilities.
Bayesian MCP
A Model Calling Protocol (MCP) server for Bayesian reasoning, inference, and belief updating. This tool enables LLMs to perform rigorous Bayesian analysis and probabilistic reasoning.
Features
🧠 Bayesian Inference: Update beliefs with new evidence using MCMC sampling
📊 Model Comparison: Compare competing models using information criteria
🔮 Predictive Inference: Generate predictions with uncertainty quantification
📈 Visualization: Create visualizations of posterior distributions
🔌 MCP Integration: Seamlessly integrate with any LLM that supports MCP
Installation
Development Installation
Clone the repository and install dependencies:
git clone https://github.com/wrenchchatrepo/bayesian-mcp.git
cd bayesian-mcp
pip install -e .
Requirements
Python 3.9+
PyMC 5.0+
ArviZ
NumPy
Matplotlib
FastAPI
Uvicorn
Quick Start
Starting the Server
# Run with default settings
python bayesian_mcp.py
# Specify host and port
python bayesian_mcp.py --host 0.0.0.0 --port 8080
# Set log level
python bayesian_mcp.py --log-level debug
The server will start and listen for MCP requests on the specified host and port.
API Usage
The Bayesian MCP server exposes several functions through its API:
1. Create Model
Create a new Bayesian model with specified variables.
# MCP Request
{
"function_name": "create_model",
"parameters": {
"model_name": "my_model",
"variables": {
"theta": {
"distribution": "normal",
"params": {"mu": 0, "sigma": 1}
},
"likelihood": {
"distribution": "normal",
"params": {"mu": "theta", "sigma": 0.5},
"observed": [0.1, 0.2, 0.3, 0.4]
}
}
}
}
2. Update Beliefs
Update model beliefs with new evidence.
# MCP Request
{
"function_name": "update_beliefs",
"parameters": {
"model_name": "my_model",
"evidence": {
"data": [0.1, 0.2, 0.3, 0.4]
},
"sample_kwargs": {
"draws": 1000,
"tune": 1000,
"chains": 2
}
}
}
3. Make Predictions
Generate predictions using the posterior distribution.
# MCP Request
{
"function_name": "predict",
"parameters": {
"model_name": "my_model",
"variables": ["theta"],
"conditions": {
"x": [1.0, 2.0, 3.0]
}
}
}
4. Compare Models
Compare multiple models using information criteria.
# MCP Request
{
"function_name": "compare_models",
"parameters": {
"model_names": ["model_1", "model_2"],
"metric": "waic"
}
}
5. Create Visualization
Generate visualizations of model posterior distributions.
# MCP Request
{
"function_name": "create_visualization",
"parameters": {
"model_name": "my_model",
"plot_type": "trace",
"variables": ["theta"]
}
}
Examples
The examples/ directory contains several examples demonstrating how to use the Bayesian MCP server:
Linear Regression
A simple linear regression example to demonstrate parameter estimation:
python examples/linear_regression.py
A/B Testing
An example of Bayesian A/B testing for conversion rates:
python examples/ab_test.py
Supported Distributions
The Bayesian engine supports the following distributions:
normal: Normal (Gaussian) distributionlognormal: Log-normal distributionbeta: Beta distributiongamma: Gamma distributionexponential: Exponential distributionuniform: Uniform distributionbernoulli: Bernoulli distributionbinomial: Binomial distributionpoisson: Poisson distributiondeterministic: Deterministic transformation
MCP Integration
This server implements the Model Calling Protocol, making it compatible with a wide range of LLMs and frameworks. To use it with your LLM:
import requests
response = requests.post("http://localhost:8000/mcp", json={
"function_name": "create_model",
"parameters": {
"model_name": "example_model",
"variables": {...}
}
})
result = response.json()
License
MIT
Credits
Based on concepts and code from the Wrench AI framework.
This server cannot be installed
Related Resources
Related MCP Servers
- AsecurityAlicenseAqualityA Model Context Protocol server that enables LLMs to perform precise numerical calculations by evaluating mathematical expressions.Last updated -1132MIT License
- -securityAlicense-qualityA Model Context Protocol server that enables LLMs to interact with databases (currently MongoDB) through natural language, supporting operations like querying, inserting, deleting documents, and running aggregation pipelines.Last updated -MIT License
- -security-license-qualityA sophisticated server that coordinates multiple LLMs (Claude, Gemini, etc.) using the Model Context Protocol to enhance reasoning capabilities through strategies like progressive deep dive and consensus-based approaches.