Frequently Asked Questions

Common questions about AiSpendTrack.

General

What is AiSpendTrack?

AiSpendTrack is a lightweight proxy that tracks your OpenAI and Anthropic API costs in real-time. You route your API calls through our proxy, we log metadata (model, tokens, cost), and you get a dashboard showing exactly where your AI budget goes.

How does it work?

You change your OpenAI/Anthropic baseURL to point to our proxy
We forward your request to OpenAI/Anthropic unchanged
We log metadata (model, tokens, cost, latency)
OpenAI/Anthropic responds
We forward the response to you unchanged
You see costs in your dashboard within 10 seconds

Do you log my prompts or responses?

No. We only log metadata:

✅ Model, tokens, cost, latency, status
❌ Your prompts, AI responses, user data

Your data passes through our proxy untouched.

Which AI providers do you support?

Currently:

✅ OpenAI (all models)
✅ Anthropic (Claude 3 family)

Coming soon:

🔜 Google Gemini
🔜 Cohere
🔜 Mistral AI

How much does it cost?

Free tier:

10,000 API calls/month tracked
7-day data retention
Basic dashboard
$0/month

Pro tier:

Unlimited API calls
90-day retention
Advanced analytics
Alerts & webhooks
$149/month

Enterprise:

Custom retention
SSO, self-hosted options
Custom pricing

Full pricing →

Setup & Integration

How long does setup take?

10 minutes on average.

Sign up (2 minutes)
Change 2 lines of code (5 minutes)
Deploy (3 minutes)
Done!

Do I need to change a lot of code?

No. Just 2 lines:


// Before
const openai = new OpenAI({ apiKey: "..." });
 
// After (2 lines)
const openai = new OpenAI({
  apiKey: "...",
  baseURL: "https://proxy.aispendtrack.com/v1",        // Line 1
  defaultHeaders: { "x-aispendtrack-key": "ask_..." }  // Line 2
});

What languages/frameworks do you support?

Officially supported:

Node.js (JavaScript, TypeScript)
Python
Go (community-supported)
Ruby (community-supported)

Works with:

Any language/framework that can make HTTP requests
All official OpenAI/Anthropic SDKs

Integration guides →

Will this break my existing code?

No. Our proxy is 100% compatible with OpenAI and Anthropic APIs. Responses are identical.

What if your service is down?

Your API calls fail. We recommend:

Implement fallback to direct OpenAI/Anthropic
Monitor our status page
We have 99.9% uptime SLA (Pro+)

Privacy & Security

Do you store my OpenAI/Anthropic API keys?

No. Your keys pass through our proxy but are never stored. They only exist in your code and OpenAI/Anthropic’s servers.

Is my data secure?

Yes:

✅ TLS 1.3 encryption in transit
✅ AES-256 encryption at rest
✅ SOC 2 Type II infrastructure
✅ Regular security audits

Security details →

Yes. We’re fully GDPR compliant:

✅ Right to access
✅ Right to deletion
✅ Right to portability
✅ Data processing agreements available

GDPR details →

Where is my data stored?

Default: United States (AWS US-East-1)

Enterprise: EU or other regions available

Can I delete my data?

Yes. Anytime:

Go to Settings → Delete Account
Confirm deletion
Data deleted within 30 days (GDPR)

Account deletion guide →

Performance

How much latency does the proxy add?

Average: <300ms

Breakdown:

Header processing: ~5ms
Database logging: ~50ms (async, doesn’t block)
Network overhead: ~50-100ms
Total: ~60ms average

Most of the “latency” is just network hops, which you’d have anyway.

Does it affect my OpenAI/Anthropic rate limits?

No. Your OpenAI/Anthropic rate limits are unchanged. You’re still using your own API key.

AiSpendTrack rate limits:

Free: 1,000 requests/hour
Pro: 10,000 requests/hour
Enterprise: Custom

Will this slow down my app?

No. The added latency (~60ms) is negligible compared to:

OpenAI GPT-4 response time: ~2-5 seconds
Claude response time: ~1-3 seconds

Users won’t notice the difference.

Pricing & Billing

Is there a free tier?

Yes! 10,000 API calls/month tracked for free.

Perfect for:

Side projects
Early-stage startups
Testing AiSpendTrack

What happens when I hit the free tier limit?

You’ll get a 429 error response. Options:

Wait until next month (resets on 1st)
Upgrade to Pro (unlimited)

We’ll email you at 50%, 75%, 90% of limit.

Can I try Pro before paying?

Yes! 14-day free trial of Pro tier.

How do I cancel?

Go to Settings → Billing
Click “Cancel Subscription”
Confirm

You keep Pro features until end of billing period.

Do you offer discounts?

Yes:

Startups (YC, etc.): 50% off first year
Non-profits: 50% off
Students: Free Pro tier
Annual billing: 2 months free

Contact sales@aispendtrack.com

Features

Can I track costs per customer?

Yes! Add customer ID to headers:


defaultHeaders: {
  "x-aispendtrack-key": "ask_...",
  "x-customer-id": "user_123"
}

Now see per-customer costs in dashboard.

Per-customer tracking guide →

Can I track costs per feature?

Yes! Add feature tag:


defaultHeaders: {
  "x-aispendtrack-key": "ask_...",
  "x-feature": "chat-support"
}

Now see per-feature costs in dashboard.

Per-feature tracking guide →

Can I set cost alerts?

Yes (Pro tier):

Daily/monthly budgets
Spike detection
Usage limit warnings
Email, Slack, webhook notifications

Alert setup guide →

Can I export my data?

Free tier: No export