Cost Analytics
AI Arbitrage Engine
Savings and quality comparisons are estimates based on published pricing and curated quality scores (methodology and reference date shown in Settings). Dollar figures assume a 60/40 input/output token split. Treat these as directional guidance, not guaranteed outcomes.
Configure Analysis
Current Model
Task Type
Monthly Token Volume
⚡
Select a model and task to find arbitrage opportunities.
AI Role Recommender
Token budgets and cost estimates are directional, based on role profiles and a 60/40 input/output split. Actual usage will vary by individual workflow, prompt length, and model behavior.
Select a Role
Industry
Role
🎯
Select a role to get model and token budget recommendations.
Data Manager
Load Demo Data
Generate sample usage data across 6 departments and 25+ employees for demonstration.
Import CSV / Excel
Accepts .csv and .xlsx. Columns: employee, department, role, model, provider, tokens_input, tokens_output, task_type, timestamp
Clear Data
Remove all imported usage data. Pricing history is preserved.
CSV Template
employee,department,role,model,provider,tokens_input,tokens_output,task_type,timestamp Alice Chen,Engineering,Senior Software Engineer,Claude Sonnet 4,Anthropic,15000,8000,code generation,2026-06-15T10:30:00 Bob Martinez,Legal,Paralegal,GPT-4o,OpenAI,8000,3000,contract review,2026-06-15T11:00:00
Cross-Module Insights
All savings figures, quality scores, and tier assignments are estimates based on stated assumptions. Treat these as a guide for where to look, not contractual projections.
Workforce ROI Calculator
All cost figures, ROI projections, and productivity estimates are directional models built on stated inputs — not guarantees. Token pricing reflects current API list rates; actual spend depends on prompt design, caching, and negotiated terms.
Hire Parameters
Role Title
Annual Base Salary
$
Benefits & OH
%
AI Model
Token Allowance / Mo
Est. Annual Value Delivered
$
Token Split — Input 30% / Output 70%
Cost Analysis — Live
$150,000
$42,000
$496
Total Fully Loaded · Annual
$192,496
$16,041
Monthly Burn
159.8%
1-Year ROI
4.6 mo
Payback
250,000 tok/mo · 75K in @ $3.00/M + 175K out @ $15.00/M = $41/mo → $496/yr
12-Month ROI Projection — Cumulative Value vs. Cost
Cumulative Loaded Cost
Cumulative Value Delivered
Breakeven Month
Ramp curve: 40 → 55 → 70 → 85 → 92 → 100% productivity over months 1–6+
Settings
AI API Key (Anthropic or OpenAI)
The "Validate with Claude" feature on the Onboarding tab sends your onboarding plan to Anthropic's or OpenAI's API (auto-detected from your key) for an AI review. This uses your own account and bills to you. Your key is stored only in this browser and is sent only to api.anthropic.com (sk-ant- keys) or api.openai.com (other sk- keys). It also powers the live Refresh Pricing search. All other features work without a key.
Data Location
Database
Data Directory
Pricing Data
Models Tracked
—
Providers
—
Jarod starts empty: use Choose Providers to pick which providers to track, or clear everything to start over. With an Anthropic or OpenAI API key saved, Refresh Pricing in the sidebar searches for live prices and lists changes in the Proposals panel for you to approve.
About
Product
Jarod — Tokenomics Arbitrage
Version
Quality scores reflect a fixed reference methodology (hand-curated for major models, price-tier-derived for dynamically discovered models). Dollar figures assume a 60/40 input/output token split. All savings and quality figures are directional estimates, not guarantees.