Polling and Credits
Teslemetry gets vehicle data in two different ways: streaming, where your vehicle sends live Fleet Telemetry to Teslemetry, and automatic polling, where Teslemetry's servers ask Tesla for a fresh snapshot on your behalf while your vehicle is online.
Automatic polling is free
Automatic polling never uses your credits, no matter how often it happens.
- A normal non-streaming (legacy) vehicle is polled for full vehicle data about every 15 minutes.
- A vehicle Tesla marks as a discounted device is polled much more often, about every 90 seconds.
"Discounted" describes how often Teslemetry can afford to poll that vehicle automatically - it does not mean automatic polls cost you 0.1 credit each. Automatic polls, discounted or not, cost you 0 credits.
What actually uses credits
Credits are used when something asks Teslemetry to fetch fresh vehicle data on demand, right now, instead of waiting for the next automatic poll or using what's already cached.
- A successful on-demand fetch normally costs 2 credits.
- On a discounted device, that drops to 0.1 credit.
- Cached reads and streamed reads are always free.
Vehicle commands (lock, climate, charging, etc.) and waking an asleep vehicle are separate, unrelated credit costs - see Handling Sleep for wake-up credits.
Cache freshness
Teslemetry's servers keep a recent copy of your vehicle's full data. For a default request, that copy is considered fresh for 20 minutes:
- If the copy is under 20 minutes old (or your vehicle is offline), a request returns it for free.
- Once it's 20 minutes or older, the first request while your vehicle is online fetches a new copy and charges credits using the rates above.
- A request that explicitly skips the cache, or Teslemetry's manual refresh endpoint, can charge credits on every single request while your vehicle is online - there's no freshness window to protect you there.
Streaming vehicles and the classic data cache
Live streaming updates keep individual fields (battery level, location, and so on) current, but they do not refresh the 20-minute full-data cache described above. That cache is only refreshed by an automatic poll or an on-demand fetch.
This matters if you're on a streaming vehicle: streaming itself is always free, but a client, app, or automation that insists on repeatedly requesting the classic full vehicle data - rather than reading the live stream or the free cache-only data - can still run into the same 20-minute cache rule and start charging credits, even though your vehicle streams. Whether a specific integration entity does this is a property of that integration, not of Teslemetry itself - if you're unsure, check its documentation, or prefer streaming/cache-only data sources for routine updates.
Summary
| Action | Costs credits? |
|---|---|
| Automatic server polling (15 min normal, 90 sec discounted) | No - always free |
| Streaming updates | No - always free |
| Cached or cache-only reads | No - always free |
| On-demand fresh data fetch, cache still fresh | No - free |
| On-demand fresh data fetch, cache stale or bypassed | Yes - 2 credits (0.1 discounted) |
| Manual refresh endpoint | Yes - every online request |