This repository implements a comprehensive framework for building intelligent Large Language Model (LLM) applications using Sberbank's GigaChat API integrated with LangChain. Designed for production-grade AI systems with a focus on Russian-language processing and advanced agent architectures.
- 🤖 Full GigaChat Integration - Native support for Sberbank's advanced Russian-language LLM
- 🔗 LangChain Ecosystem - Modular pipelines, chains, and agents
- 📚 Advanced RAG Systems - Retrieval-Augmented Generation with multiple vector stores
- 🔄 Multi-Agent Architectures - Collaborative intelligent agent systems
- 🌐 API-First Design - RESTful endpoints with FastAPI
- 🐳 Production Ready - Docker, monitoring, logging, and deployment scripts
- 🔒 Security Focused - Built-in rate limiting, authentication, and compliance features
- Python 3.9+
- GigaChat API access (from developers.sber.ru)
- Git
# Clone repository
git clone <your-repo-url>
cd llms-framework
# Create virtual environment
python -m venv venv
source venv/bin/activate # On Windows: venv\Scripts\activate
# Install dependencies
pip install -r requirements.txt
# Configure environment
cp .env.example .env
# Edit .env with your GigaChat API credentials