1. Select Agent Scenario Preset
2. Select Core LLM Model
3. Define Run Parameters
4 turns
Number of sequential prompts/actions the agent makes.How many times your team runs this type of agent task monthly.
Context Strategy Mode
Agents keep previous file edits and instructions in chat history.
40%
Represents how much extra content (retrieved files, tools outputs, terminal diagnostics) is appended to each agent step.
Includes current file contexts, diffs, and codebase guidelines.
Tokens generated by the LLM (code writes, text planning).
Compute / Execution Cost (10 cr / GB-hour)
6 seconds
Average time spent executing shell commands/searches per turn.High memory configurations consume credits proportionally faster.
Production Deploys (15 cr / deploy)
Successful production launches triggered by the agent.
1
4. Prompt Token Size Estimator
Paste your planned prompt below to calculate its approximate token length and apply it directly as your Input Token baseline.
Real-time Calculations
SINGLE RUN
101credits
Worth $0.56 in model tokens
EST. MONTHLY
1,509credits
Worth $8.38 in model tokens
Credit Breakdown
AI Inference (Tokens)
85.5 crCompute (GB-Hours)
0.067 crProduction Deploys
15 crPlan Compatibility
Free (300 cr/mo)
Exceeded100% of allowance usedOver limit
Personal (1,000 cr/mo)
Exceeded100% of allowance usedOver limit
Pro (3,000 cr/mo)
Within Budget50.3% of allowance used+1,491 surplus
Pro Tip: Paid plans can enable auto-recharge. Overage credits are purchased automatically in blocks of 1,500 credits for $10 USD on Pro plans.