Observed arrival · 2026-09-23
TokenPolice puts a stop rule before the LLM call
TokenPolice meters AI calls inside an app and lets teams set rules to block, reroute, or flag spending before a request reaches a provider.
○Open
⊠Login
$Paid
†Ads
✦Pretty
●Pro
◎Niche
◉Human
⚑Risk
ƒJS
- For
- Teams running LLM calls inside Node or Python apps
- Worth noticing
- The page lists 13 providers and seven frameworks with direct metering, plus 14 more through OpenAI-compatible endpoints.
Field notes
The page describes integrations for Node and Python, with model calls attributed to a run and usage broken down by user, plan, feature, session, and workflow. It says rules can cover different call types, including embeddings, image generation, speech, video, and OCR, rather than treating every request as chat. The site also says teams can describe prices for fine-tuned, private, or negotiated-rate models in their own terms.
Observed signals
Read the marks
Editorial observations of this landing page, not a rating.
○OpenPublic substance visible
$PaidCommerce or pricing visible
✦PrettyNotable craft visible
◎NicheUnusually specific use
ƒJavaScriptBrowser-side code central
One card from the complete issue