Cost Tracking

Chatelope helps you track API costs across all your AI usage, whether from imported conversations or the built-in chat feature.

Why Track Costs?

AI API costs can add up quickly:

• GPT-4 costs more than GPT-3.5
• Long conversations use more tokens
• Image generation has per-image costs
• Different providers have different pricing

Without tracking, it's easy to overspend.

Setting Up Cost Tracking

Configure Providers

1. Go to Settings > AI Settings
2. Add your API keys
3. Chatelope automatically tracks usage

Import Historical Data

When you import conversations:

1. Token counts are estimated from message length
2. Model information is extracted when available
3. Costs are calculated based on known pricing

Analytics Dashboard

Go to Analytics to see your usage.

Overview

• Total spend (all time / this month)
• Spend by provider
• Trend over time

By Provider

See costs for each provider:

By Model

Breakdown by specific model:

By Time Period

View usage over time:

• Daily breakdown
• Weekly trends
• Monthly comparison

Budget Alerts

Set spending limits:

1. Go to Settings > Usage & Billing
2. Click Set Budget
3. Enter monthly limit
4. Choose alert threshold (e.g., 80%)

Alert Types

• Email - Get notified by email
• In-app - Banner in Chatelope
• Both - Multiple notifications

When Triggered

At your threshold:

1. Alert is sent
2. Optional: Pause API usage
3. Review in Analytics

Cost Optimization

Model Selection

Use cheaper models when appropriate:

Conversation Length

Manage conversation length:

• Start new conversations for new topics
• Don't include unnecessary context
• Use system prompts efficiently

Caching

Enable response caching:

1. Go to Settings > AI Settings
2. Enable Response Cache
3. Identical prompts use cached responses

Reports

Generate Reports

Create spending reports:

1. Go to Analytics > Reports
2. Select date range
3. Choose breakdown (by provider, model, project)
4. Export as CSV or PDF

Schedule Reports

Set up automatic reports:

1. Go to Analytics > Scheduled Reports
2. Choose frequency (weekly/monthly)
3. Select report format
4. Enter email addresses

Team Usage

If sharing API keys:

Per-User Tracking

Track usage by user:

• Each user's spending
• Model preferences
• Usage patterns

Cost Allocation

Allocate costs to projects:

• Tag conversations with project names
• View spending by tag
• Export for accounting

Tips

Before the Month Ends

1. Check current spending
2. Review high-cost conversations
3. Adjust model selection if needed

Reduce Waste

1. Don't retry failed requests unnecessarily
2. Use streaming to see partial responses
3. Cancel unwanted generations early

Compare Providers

Use multiple providers to find the best value:

• Try the same prompt on different models
• Compare quality vs. cost
• Choose provider per use case