CellSem LLM Client Documentation

Contents

• Installation
  • Requirements
  • Quick Installation
  • Environment Setup
  • Verify Installation
  • Development Installation
  • Docker Installation (Optional)
  • Troubleshooting
• Quick Start Guide
  • Basic Usage
  • Working with System Messages
  • Provider-Specific Features
  • Advanced Usage
  • Best Practices
  • Example Applications
  • Next Steps
• Schema Enforcement (JSON-First)
  • Core Concepts
  • Quick Example (JSON-First)
  • Schema Inputs
  • Under the Hood
  • Notes
• Cost Tracking and Estimation
  • Estimated Costs (Per Request)
  • Actual Usage (Key-Level)
  • Rates and Updates
• Contributing
  • Development Philosophy
  • Getting Started
  • Development Workflow
  • Contributing Guidelines
  • Development Commands
  • Architecture Guidelines
  • Getting Help
  • Recognition
• API Reference
  • cellsem_llm_client

Welcome

CellSem LLM Client is a flexible LLM client with multi-provider support, built using LiteLLM + Pydantic for seamless integration across OpenAI, Anthropic, and other providers.

Key Features

Multi-Provider Support

Seamless switching between OpenAI, Anthropic, and other LiteLLM-supported providers

Type Safety

Full MyPy type checking with strict configuration for reliable code

Comprehensive Testing

Dual testing strategy: real APIs for local development, mocks for CI

Environment Configuration

Automatic API key loading and configuration from environment variables

Quick Links

• Installation - Get started with installation and setup

• Quick Start Guide - Basic usage examples and workflows

• cellsem_llm_client - Complete API reference documentation

• Contributing - Development guidelines and TDD workflow

Current Status

Phase 2 - Complete: Core multi-provider functionality with comprehensive testing

Phase 3 - In Development: Advanced features including:

• Token tracking & cost monitoring

• JSON schema compliance

• File attachment support

• AI-powered model recommendations

See our GitHub repository for the latest updates and ROADMAP.md for detailed implementation plans.