Plans, messages and limits

• Everyday messages

  The everyday models (Claude Haiku 5.5, GPT-6 Luna, DeepSeek V4.1 Flash, and GLM 5.3 Flash) use one everyday message per message, from a daily count on every paid plan: 60 a day on Pro, 120 on Max, 200 on Ultra, 200 on Studio.

• One monthly allowance

  Every other model draws on one monthly allowance, shown as messages: the app tells you how many you have left on the model you pick, before you send. Bigger models use it faster. Paid tools draw on it too.

• Free's trial

  Free is a one-time trial of 5 messages on every model but Claude Fable 5.1 and GPT-6 Astra (one of them can be on Claude Opus 5.5). Each message uses 1 of them (2 or 4 in a long chat), and they don't refill.

• Every day

  Everyday messages refill at 00:00 UTC every day on every paid plan.

• Every billing period

  The monthly allowance renews on your billing date each month; unused messages don't carry over. The sidebar and Settings → Billing show what's left and when it refills.

• On Free

  Free's 5 trial messages don't refill: the trial is once.

Every paid plan includes up to a fixed amount of AI cost each billing period: Pro $7.50, Max $20, Ultra $42 and Studio $85 of AI cost a billing period, at list prices; a $10 top-up adds $4 of room in the period it's bought in. At 80% we tell you; at the limit everything pauses until the period renews or you top up. Using every message on your plan at typical sizes stays under it; very large messages and heavy research use it faster. Settings → Billing shows where you stand. Free has no fair-use limit: its 5 messages are the limit. How the limit is set.

• Long chats: once the conversation sent to the model passes 64k tokens, each reply counts as 2; past 128k, as 4; past 200k the chat is full. Your memory and a project's instructions and files count toward a chat's length. The composer says so before you send: “Long chat — each reply counts as 2. Start a new chat to save messages.”

• Reply length: a reply can be up to 8k tokens on Free (Pro 8k, Max 16k, Ultra 16k, Studio 16k). At the limit it stops and offers Continue, which is a new message at the same price.

• Replies an hour: every plan has a soft cap of 100.

• Web search: on Free it works inside your 5 trial messages, up to 3 searches a day (reading a link counts as one); on a paid plan each search counts as one more message of the model in use.

On a paid plan you can add more from Settings → Billing: $10 adds 25 messages on Claude Fable 5.1 or GPT-6 Astra; 50 on Claude Opus 5.5; 100 on Claude Sonnet 5.5-class models; or 200 on Claude Haiku 4.5-class models (+ applicable tax, shown before you pay). Your plan's own allowance is spent first, then top-ups, oldest first. Each top-up lasts 12 months, and adds $4 of fair-use room in the period it's bought in. If you cancel, you can keep using your top-ups for 30 days after your plan ends, for messages only (fast models count 1; web search needs a plan), within your last billing period's fair-use limit. Anything left then waits until 12 months after purchase, in case you come back.

Plans are monthly. Change or cancel from pricing or Settings → Billing. An upgrade starts right away: you pay the difference for the rest of the period, and this period's allowance grows by the same share. A downgrade or a cancellation takes effect at the end of the period you paid for.

Every number and limit, and how we set them: How pricing works.