Stop guessing your
AI API costs

Tokonomics sits between your app and any LLM provider. Every call is tracked, every dollar counted, and you're alerted before you blow your budget.

Start free — no credit card How it works
Tokonomics PROXY
# Before — direct call to OpenAI
POST api.openai.com/v1/chat/completions

# After — one URL change
POST tokonomics.ca/proxy/openai/
chat/completions

# Tokonomics captures silently
✓ tokens: 10 in / 2 out
✓ cost_usd: $0.00000270
✓ alerts checked — on track
3Lines to integrate
$49/moStarter plan
<1msProxy overhead
ALLLLM providers
Features
Everything you need
to control AI costs

No more surprises at the end of the month.

🔁
Transparent proxy
Drop-in replacement for OpenAI, Anthropic, or DeepSeek. Change one URL — your app code stays exactly the same.
Learn more →
💰
Real-time cost tracking
Every token priced the moment the response arrives. 8-decimal USD precision using bcmath — never floats for money.
Learn more →
🚨
Budget alerts
Set thresholds at 50%, 80%, 100% of your monthly budget. Notified by email or webhook before you exceed your limit.
Learn more →
📊
Analytics dashboard
Spend by model, team, and feature. Daily trends for 90 days. Full REST API for custom reporting and integrations.
Learn more →
🏷️
Tag-based attribution
Tag every call with {"team":"growth"}. Break down costs by any dimension that matters.
Learn more →
🔑
Multiple API keys
Issue keys per team or service. Revoke instantly. Keys stored as SHA-256 hashes — plaintext shown only once at creation.
Learn more →
$0
Surprise bills
8
Decimal USD precision
<1ms
Proxy overhead
ALL
LLM providers
How it works
From request to insight in milliseconds

Tokonomics intercepts every LLM call transparently — your users never notice a difference.

01
Point your app at Tokonomics
Replace the provider base URL. Add your mk_ API key as Bearer token. One line change.
02
Tokonomics forwards in real-time
Your request is forwarded to OpenAI, Anthropic, or DeepSeek. Response is streamed back — zero buffering, zero latency penalty.
03
Usage captured silently
Tokens and cost extracted from the provider response and recorded in the database without slowing you down.
04
Alerts fire automatically
When spend crosses a threshold, Tokonomics fires a webhook or email. Stay in control without checking dashboards.
Pricing
Simple, transparent pricing

No per-request fees. No token surcharges. Flat monthly subscription.

Starter
$49
per month
  • ✓ 1 seat
  • ✓ 3 budget alerts
  • ✓ 30-day data retention
  • ✓ OpenAI · Anthropic · DeepSeek
  • ✓ Email alerts
  • ✓ REST analytics API
Get started
Pro — Most Popular
$99
per month
  • ✓ 5 seats
  • ✓ Unlimited alerts
  • ✓ 90-day data retention
  • ✓ Email + webhook alerts
  • ✓ Hard spending caps
  • ✓ Tag-based analytics
Start Pro trial
Enterprise
$299
per month
  • ✓ Unlimited seats
  • ✓ 12-month data retention
  • ✓ SSO (SAML/OIDC)
  • ✓ White-label & custom domain
  • ✓ Priority support
  • ✓ SLA guarantee
Contact sales
FAQ
Frequently asked questions
Tokonomics adds less than 1ms of overhead. It streams the provider response back in real-time without buffering — your users will not notice any difference.
No. Tokonomics only records token counts, cost, latency, model, provider, and your custom tags. The content of your prompts and completions is never stored.
Currently OpenAI, Anthropic, and DeepSeek. Any OpenAI-compatible provider also works out of the box. More native integrations are on the roadmap.
Tokonomics API keys are stored as SHA-256 hashes — the plaintext is shown exactly once at creation and never stored. Your provider keys are stored in environment variables and never exposed.
Yes. Tokonomics is a standard HTTP proxy — you only need to change the base URL in your existing LLM client. It works with any language or framework.
On Starter you receive email alerts at configured thresholds. On Pro and above you can also set hard caps that block proxy requests at the API level once the cap is hit.

Ready to stop guessing your AI costs?

Join PHP teams who know exactly what they spend — before it hurts.

Create free account →