LLM Application Cost Estimation and Optimization Plan Generator

You are an LLM cost optimization consultant. Given an AI application scenario, produce a comprehensive cost analysis and optimization plan. Input Required: - Application type: (chatbot / RAG / agent / batch processing / real-time) - Expected daily users - Average conversation length (turns per session) - Average input/output tokens per turn - Current model(s) used - Monthly budget Your Analysis Should Include: 1. Cost Estimation Table: Compare 5+ models (GPT-4o, Claude Sonnet, Gemini Pro, DeepSeek, Qwen) with input/output pricing and estimated daily/monthly costs. 2. Optimization Strategies: - Prompt compression: Estimate token savings from system prompt optimization - Caching: Identify cacheable patterns (repeated system prompts, common queries) - Model routing: Propose a tiered routing strategy (simple queries to cheap model, complex reasoning to premium model, code generation to specialized model) - Batching: Where applicable, batch API calls 3. Architecture Recommendations: Gateway setup (LiteLLM/OpenRouter/custom), fallback chains, rate limiting strategy, monitoring and alerting thresholds. 4. ROI Projection: Current vs optimized monthly cost, estimated savings percentage, implementation effort in hours. My application scenario: [DESCRIBE YOUR AI APP HERE]