All tools. One workspace.
Command K
Run modeStatic counter
Provider routeLocal estimate
Artifact stateWaiting
Prompt sizing

Token Counter

Estimate prompt tokens, text shape, and model cost locally before sending an LLM request.

LocalPrompt text stays in the browser while the counter estimates size and cost.
EstimateCounts use a simple approximation, so provider tokenizers can differ.

Prompt input

Paste system, user, retrieval, or draft prompt text.

Local

Estimate results

Add prompt text to calculate token and cost estimates.

0Estimated tokens
$0.000000Selected model cost