does 150 requests mean 150 msgs i can send? doesn't matter how many tokens it uses?
#how usage work?
atomic pike
lunar ivyBOT
Hi @atomic pike. I'm kapa.ai, your friendly support bot. I will try to answer your question by searching through the knowledge base. :wave:
Warp's usage system is based on "AI requests," but it's not as simple as "1 message = 1 request." Each time you interact with Warp AI (e.g., send a prompt, generate code, or use command suggestions), you consume at least one AI request. However, more complex interactions—such as those involving large codebases, multiple tool calls, or lots of context—can consume multiple requests for a single message. The number of requests used is non-deterministic and depends on factors like the model chosen, the complexity of the task, and the amount of context or data attached to your prompt. So, 150 requests does not necessarily mean you can send exactly 150 messages; some messages may use more than one request, especially if they are complex or involve a lot of data or steps.
Tokens are a separate concept. While tokens (chunks of text processed by the AI) influence how many requests are consumed, you do not need to track tokens directly—Warp abstracts this away, and your plan is limited by the number of requests, not tokens. However, if you hit a token limit (which is rare and usually only happens with very large or complex prompts), you may see a separate error message about exceeding the token limit. But for most users, you only need to monitor your AI request usage, not tokens How AI requests are calculated What counts as an AI token?.
Summary:
- 150 requests ≠ 150 messages.
- Each message may use 1 or more requests, depending on complexity and context.
- You do not need to track tokens; just monitor your request usage in Warp.