Spaces:
Runtime error
Runtime error
| title: Dispatch AI — AI Cost Calculator | |
| emoji: 💰 | |
| colorFrom: blue | |
| colorTo: indigo | |
| sdk: gradio | |
| sdk_version: "4.44.1" | |
| app_file: app.py | |
| pinned: true | |
| license: other | |
| tags: | |
| - cost-calculator | |
| - on-device | |
| - savings | |
| - co2 | |
| - dispatch-ai | |
| # Dispatch AI — AI Cost Calculator 💰 | |
| Input your cloud API usage (tokens/month) → see cost savings by switching to on-device AI inference. Shows breakeven point, 1-year savings, and CO2 reduction. | |
| Built by **Dispatch AI (FZE)** — UAE · License 10818. | |
| ## Features | |
| - 💵 **Total savings** — cloud vs on-device over your projection period | |
| - ⏱️ **Breakeven point** — when on-device becomes cheaper than cloud | |
| - 🌍 **CO2 reduction** — environmental impact of switching | |
| - 📊 **Side-by-side comparison** — monthly cost, total cost, CO2 emissions | |
| - 🌳 **Tree equivalent** — how many trees your CO2 savings represent | |
| - ⚡ **Capacity check** — can your phone fleet handle the token volume? | |
| - 📋 **8 cloud providers** — GPT-4o, GPT-4o-mini, Claude 3.5 Sonnet/Haiku, Llama, Mistral, Gemini | |
| - 🌙 **Dark theme** (#0A0F1A bg, #1FE0E6 cyan accent) | |
| ## How It Works | |
| 1. Enter your monthly input/output token usage (in millions) | |
| 2. Pick your current cloud provider | |
| 3. Set how many phones you'd deploy | |
| 4. Choose projection period (1-36 months) | |
| 5. Click **Calculate Savings** | |
| ## Assumptions | |
| - Phone: Samsung S20 FE ($400, 24-month lifespan) | |
| - Inference: 16 t/s (Qwen2.5-1.5B Q4_K_M on SD865) | |
| - Electricity: ~$2.50/month/phone | |
| - CO2: 0.4 kg/1M tokens (cloud) vs 0.384 kg/month (phone) | |
| ## License | |
| © 2026 Dispatch AI FZE — Sharjah, UAE · License 10818. All rights reserved. | |