Incorrect responses still using tokens?

[pawlean (Pauline P. Narvas)]

Hi, Jabber! Welcome to the Vercel Community.

Yes, tokens are consumed for every AI request regardless of the quality of the response. When you send a prompt to v0, tokens are counted for both your input (the prompt) and the AI’s output (the response), even if that response doesn’t meet your expectations.

I understand your frustration with the new token-based pricing. The v0 token system is designed to provide more granular usage tracking, but I recognize this change impacts how you use the platform.

We appreciate the feedback! Feel free to drop it in the main thread as well, if you’d like:

Updated v0 pricing

v0 has a new usage-based pricing model: Draw down from a monthly credit balance based on tokens used per message Access detailed information on usage and limits for each plan Purchase additional credits on paid plans, so you can always use v0 For existing users, you will transition to the new pricing structure at the start of your next billing period. The plan prices remain unchanged at $20 for Premium and $30/seat for Team. You can opt-in immediately at: https://v0.dev/chat/settings/billi…