Free 5 Minute Assessment
Based on original research simulating nearly 240,000 LLM users, this assessment identifies which of five behavioral archetypes best describe your AI usage pattern.
Some people look for a LLM cost calculator or LLM model cost comparison tool to understand and control their Claude Code, Cursor, Codex or other LLM API costs. But the biggest levers are behavioral, not technical.
This assessment reveals what behaviors drive your LLM costs over time and shows you exactly where the savings are. It takes less than 5 minutes to complete.
Complete the assessment by answering XX questions. During the assessment, you'll provide information about how you use AI for common tasks.
You don't have to be an AI expert to take the assessment. You'll spend time thinking about how you start AI sessions, manage AI during them, and work with the AI when it gets something wrong.
The assessment is for everyone from coders to non-technical founders working with AI to launch products and services — and everyone in between. You also don't have to be an expert AI user to take the assessment.
It will take less than 5 minutes to complete
Your personal information is not stored with survey results and is only used to deliver reports and LLM cost reduction coaching if you opt in to receive these benefits.
Assessment
Your Archetype
Achieves the lowest cost per task through clean workflow habits and disciplined model selection.
A 13-question assessment based on research simulating nearly 240,000 LLM users. It identifies your behavioral archetype, one of five distinct patterns: LLM Whisperer, Eager Conversationalist, Stubborn Economizer, Precise Spender, or Free Spender, and shows how your habits affect your LLM costs.
Many people try to reduce their LLM cost by switching models. Others use a LLM cost calculator to predict costs. But the data shows that model choice alone explains only a fraction of the cost picture. The biggest cost drivers are behavioral: retries, cache usage, context bloat, and prompt overhead. Together, these factors have a much larger impact on cost than what LLM model you pick.
This assessment was created to make those hidden cost drivers visible to you. It reveals the specific behavioral patterns that explain why two people using the same models can have dramatically different LLM api cost outcomes. It's an LLM cost optimization tool for people who want to fix the root cause, not just the symptoms.
This assessment is grounded in Anthropic's analysis of 400,000 Claude Code sessions and our own simulation of nearly 240,000 simulated LLM users of Claude Code, Codex, Cursor and other AI platforms. It shows that understanding your own behavioral patterns is the highest-leverage move you can make toward achieving LLM cost optimization success.
It measures five behavioral variables that drive LLM costs: how often you retry failed tasks (the #1 cost driver), how you handle session context and caching, which model tier you default to, how much background context you include, and how well your beliefs about LLM usage align with research findings.
Each variable has a measurable impact on your cost per successful task. For example, users who retry frequently pay significantly more for the same outcomes.
The assessment is based on original research simulating nearly 240,000 LLM users to identify behavioral patterns that correlate with cost efficiency. The five archetypes and their associated cost profiles were derived from simulation data showing how different behaviors shape cost outcomes.
The archetypes are behavioral patterns with measurable cost impacts. Your archetype tells you what behaviors you tend toward, and reveals your biggest cost-reduction opportunities.
The assessment is free and takes about 5 minutes to complete.
LLM Whisperer: Achieves the lowest cost per task through clean workflow habits and disciplined model selection. Gets it right the first time with fewer turns, streamlined context, and high cache utilization.
Eager Conversationalist: Keeps conversations active and achieves high cache coverage, but verbose sessions accumulate context bloat that dilutes the caching advantage.
Stubborn Economizer: Defaults to economy-tier models but loses savings through repeated, non-productive turns when sessions go wrong.
Precise Spender: Writes well-constructed prompts with low retry rates, but defaults to flagship models and skips caching, erasing efficiency gains.
Free Spender: Combines flagship model pricing with high retry rates, producing the highest cost per successful task of any profile.
After completing the assessment you'll get your behavioral archetype, efficiency score, grade, and biggest cost-reduction lever. If you opt in to get your personalized report, you'll also receive two personalized 90-day coaching programs tailored to your specific patterns.
The available coaching programs include: Retry Coach, Cache Coach, Model Coach, Prompt Engineer Coach, and Token Economics 101. The assessment system will recommend the two most relevant programs based on the behavioral habits that are driving up your LLM costs the most.
The token economics course is always included.
The behavioral patterns and cost drivers revealed by this assessment are grounded in original research: a simulation of nearly 240,000 LLM users of AI platforms like Claude Code, Codex, Cursor, and others. Read the report to understand the behavioral archetypes and how they shape LLM costs.
