Skip to content

πŸ€– Explain with AI

Token Usage

AI Studio provides comprehensive token usage tracking across all pods and AI models. Monitor consumption, set budgets, and optimize costs.

Overview

Token usage tracking includes:

  • Per-Model Breakdown β€” See which models consume the most tokens
  • Per-Pod Tracking β€” Monitor consumption by project
  • Per-Task Detail β€” Understand token cost of individual tasks
  • Audit Log β€” Complete history of all token consumption
  • Budget Controls β€” Set limits and alerts

Token Consumption by Model

Different AI models have different token costs:

Model Input Cost Output Cost Best For
Claude 3.5 Sonnet Higher Higher Complex reasoning, architecture
GPT-4o Medium Medium General tasks, documentation
Gemini 1.5 Pro Medium Medium Large context, multi-file analysis
DeepSeek Coder Lower Lower Cost-efficient code generation

Viewing Model Breakdown

  1. Navigate to AI Studio
  2. Click Token Usage
  3. View the model breakdown chart:
Token Usage by Model (This Month)
β”Œβ”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”
β”‚ Claude 3.5 Sonnet  β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ 45%         β”‚
β”‚ GPT-4o             β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ 30%           β”‚
β”‚ Gemini 1.5 Pro     β–ˆβ–ˆβ–ˆβ–ˆ 15%             β”‚
β”‚ DeepSeek Coder     β–ˆβ–ˆ 10%               β”‚
β””β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”˜

Per-Pod Usage

Each pod tracks its own token consumption:

Pod Tokens Used Budget Remaining
API Migration 125,000 200,000 75,000
Mobile App 89,000 150,000 61,000
Documentation 34,000 50,000 16,000

Pod Usage Detail

Click on a pod to see:

  • Total tokens consumed
  • Breakdown by model
  • Breakdown by task
  • Daily consumption trend
  • Remaining budget

Per-Task Usage

Individual tasks show their token cost:

Task Model Input Tokens Output Tokens Total
Design API schema Claude 3.5 2,500 4,200 6,700
Write unit tests GPT-4o 1,800 3,100 4,900
Refactor module DeepSeek 3,200 2,800 6,000

Audit Log

The token audit log records every AI operation:

Timestamp Pod Task Model Input Output Total
2026-06-08 14:32 API Migration Design schema Claude 3.5 2,500 4,200 6,700
2026-06-08 14:28 Mobile App Write tests GPT-4o 1,800 3,100 4,900
2026-06-08 14:15 API Migration Refactor DeepSeek 3,200 2,800 6,000

Filtering the Audit Log

Filter by:

  • Date Range β€” Specific time period
  • Pod β€” Single pod or all pods
  • Model β€” Specific AI model
  • Task β€” Individual task

Exporting Audit Data

  1. Navigate to the audit log
  2. Apply filters
  3. Click Export
  4. Choose format: CSV or JSON
  5. Download the report

Subscription Plans

AI Studio offers three subscription tiers:

Solo

Feature Details
Pods 1 active pod
Monthly Tokens 500,000 tokens
Models All models available
File Storage 5 GB
Team Members 1 (yourself)
Support Community support

Starter

Feature Details
Pods 5 active pods
Monthly Tokens 2,000,000 tokens
Models All models available
File Storage 25 GB
Team Members Up to 5
Support Email support

Pro

Feature Details
Pods Unlimited pods
Monthly Tokens 10,000,000 tokens
Models All models + priority access
File Storage 100 GB
Team Members Unlimited
Support Priority support + SLA

Managing Your Subscription

  1. Navigate to AI Studio β†’ Subscription
  2. View current plan details
  3. Click Upgrade or Downgrade to change plans
  4. Changes take effect at the next billing cycle

Budget Controls

Setting a Monthly Budget

  1. Navigate to AI Studio β†’ Token Usage
  2. Click Set Budget
  3. Enter your monthly token limit
  4. Choose action when limit is reached:
  5. Alert Only β€” Continue but notify administrators
  6. Pause Pods β€” Suspend all AI processing
  7. Block New Tasks β€” Prevent new task creation
  8. Click Save

Budget Alerts

Configure alerts at threshold levels:

Threshold Action
50% Email notification
75% Email + dashboard warning
90% Email + dashboard warning + Slack alert
100% Execute budget action (alert/pause/block)

Token Top-Ups

If you need more tokens mid-cycle:

  1. Navigate to AI Studio β†’ Subscription
  2. Click Buy More Tokens
  3. Select top-up amount:
  4. 100,000 tokens
  5. 500,000 tokens
  6. 1,000,000 tokens
  7. Custom amount
  8. Confirm purchase

Top-up tokens are added immediately and don’t affect your next billing cycle.

Cost Optimization

Tips for Reducing Token Usage

  1. Use DeepSeek for routine tasks β€” Lower cost for standard code generation
  2. Batch similar tasks β€” Combine related work to reduce context overhead
  3. Provide clear requirements β€” Reduces iteration and rework
  4. Limit file context β€” Only attach necessary files to pods
  5. Review before re-running β€” Check if existing output can be modified

Model Selection Guide

Task Type Recommended Model Cost
Simple code generation DeepSeek Coder $
Documentation writing GPT-4o $$
Architecture design Claude 3.5 Sonnet $$$
Large file analysis Gemini 1.5 Pro $$
Code refactoring DeepSeek Coder $
Complex debugging Claude 3.5 Sonnet $$$

Reporting

Usage Reports

Available reports:

Report Description
Daily Summary Token usage per day
Weekly Trend Week-over-week comparison
Monthly Report Full monthly breakdown
Pod Comparison Usage across all pods
Model Analysis Cost-effectiveness by model

Scheduled Reports

Set up automated report delivery:

  1. Click Schedule Report
  2. Choose report type
  3. Set frequency (daily, weekly, monthly)
  4. Add recipient emails
  5. Click Save

Best Practices

  1. Monitor daily β€” Check token usage at the start of each day
  2. Set budgets early β€” Establish limits before usage grows
  3. Review audit logs β€” Identify unexpected consumption patterns
  4. Right-size models β€” Use cost-efficient models when appropriate
  5. Plan token needs β€” Estimate requirements before starting large projects

What’s Next

  • Pods β€” Manage AI project teams
  • AI Gateway β€” Learn about the underlying infrastructure

Still questions? Ask on Discord or explore tutorials