neuralnode 1.0.0

A comprehensive Python framework for building AI agents with support for both cloud-based LLM APIs and local models

NeuralNode

A next-generation Python framework for building, running, and managing Large Language Models (LLMs) and AI Agents, locally or via cloud providers.

The smarter, simpler, and more powerful alternative to LangChain.

Features

• Unified LLM Interface - One API for all LLMs (local or cloud)
• Multi-Provider Support - OpenAI, Anthropic, Google, Ollama, HuggingFace
• Streaming & Async - Real-time responses with full async support
• Function Calling - Native tool use with structured outputs
• Local AI Engine - Auto-download, quantization, hardware optimization
• AI Agents - Convert any LLM into an autonomous agent
• Browser Integration - Chrome/Edge automation built-in
• RAG Built-in - PDFs, documents, vector stores, semantic search
• Real-Time Context - Automatic time/date injection
• Memory Management - Conversation history, context-aware chaining

Quick Start

pip install neuralnode

from neuralnode import NeuralNode

# Auto-detect best provider
ai = NeuralNode(provider="auto")

# Simple chat
response = ai.chat("Explain quantum computing")
print(response.text)

# With streaming
for chunk in ai.chat("Tell me a story", stream=True):
    print(chunk.text, end="")

# Async support
response = await ai.achat("Hello!")

# Create an agent
agent = ai.agent()
result = agent.run("Find the latest news about AI")

Installation

# Base install
pip install neuralnode

# With specific providers
pip install neuralnode[openai,anthropic]

# Full install with all features
pip install neuralnode[all]

Architecture

neuralnode/
├── core/           # Base interfaces and types
├── providers/      # LLM provider implementations
├── agents/         # Agent system and orchestration
├── tools/          # Tool integrations (browser, APIs)
├── rag/            # Retrieval-Augmented Generation
├── memory/         # Conversation and context management
├── local/          # Local model runtime and hardware
└── utils/          # Utilities and helpers

License

MIT License - see LICENSE file for details.