How usage limits work
What each plan includes, how usage is measured, and what happens when you reach your limit.
What each plan includes
- Free — one free answer from a premium model when you create an account (on top of the one you get as a guest), plus saved chat history, projects and your file library. Chatting beyond that needs Pro.
- Pro — unlimited everyday chat with Xerva Instant (subject to our Fair usage policy), plus every premium model, image generation, web search, Deep Research and file analysis within a monthly usage allowance and a daily cap.
- 7-day trial — full Pro access with a smaller allowance for the whole trial. See How the trial works.
Pro is sold one way: the 7-day trial for €1.00, which then renews at €29.99 every month until you cancel.
How your allowance is applied
Your allowance renews each month. Inside it there is a daily cap, so one busy day can't use up the whole month.
During the trial, one allowance covers the whole seven days.
Xerva Instant never counts toward this allowance. On paid plans it has its own generous daily fair-use limit instead.
Checking where you stand
Open Settings → Plans and usage to see how much of your monthly allowance, today's cap and Instant fair use is left, and when each resets. The model picker also warns you when you're close to the limit.
What counts toward your usage
Usage is measured by what each reply takes to run, so not every message costs the same:
- Model — more capable models use more per reply. Claude Fable 5 and Opus 5 use the most; Gemini, DeepSeek and Kimi the least.
- Conversation length — long chats re-send their history with every message, so later replies use more. A new chat per topic makes your allowance go further.
- Files and tools — attachments, web search and Deep Research add to what a request uses.
- Images — each generated image counts by the model that made it.
How many messages does a plan buy?
Estimates for a typical message on Pro, model by model:
| Model | Pro |
|---|---|
| Claude Fable 5 | ≈ 124 / mo≈ 8 / day |
| Claude Opus 5 | ≈ 248 / mo≈ 16 / day |
| Claude Sonnet 5 | ≈ 620 / mo≈ 41 / day |
| Claude Opus 4.8 | ≈ 82 / mo≈ 5 / day |
| GPT 5.6 Sol | ≈ 314 / mo≈ 20 / day |
| GPT 5.6 Terra | ≈ 544 / mo≈ 36 / day |
| GPT 5.5 | ≈ 353 / mo≈ 23 / day |
| GPT 5.4 | ≈ 430 / mo≈ 28 / day |
| Gemini 3.1 Pro | ≈ 695 / mo≈ 46 / day |
| Gemini | ≈ 4,815 / mo≈ 321 / day |
| Grok 4 | ≈ 408 / mo≈ 27 / day |
| DeepSeek V4 Pro | ≈ 2,423 / mo≈ 161 / day |
| Kimi K2 | ≈ 1,600 / mo≈ 106 / day |
| Qwen 3 Max | ≈ 754 / mo≈ 50 / day |
Estimates only — real counts vary with conversation length, attachments and web search. The daily figure is the cap for a single day; the monthly figure is what the full month's allowance buys.
Image generation
Images per plan, one image per generation:
| Model | Pro |
|---|---|
| Nano Banana | ≈ 150 / mo≈ 10 / day |
| Nano Banana 2 | ≈ 85 / mo≈ 5 / day |
| Nano Banana 2 Lite | ≈ 171 / mo≈ 11 / day |
| Nano Banana 2.1 | ≈ 171 / mo≈ 11 / day |
| Nano Banana Pro | ≈ 42 / mo≈ 2 / day |
| GPT Image 2 | ≈ 33 / mo≈ 2 / day |
| GPT Image 1.5 | ≈ 33 / mo≈ 2 / day |
| GPT Image Mini | ≈ 120 / mo≈ 8 / day |
| FLUX.2 Pro | ≈ 120 / mo≈ 8 / day |
| FLUX.2 Flex | ≈ 200 / mo≈ 13 / day |
| FLUX.2 Max | ≈ 75 / mo≈ 5 / day |
What happens at the limit
- Premium allowance used up (monthly, or today's cap) — you're switched to Xerva Instant with a notice in the chat, and can keep chatting right away. Image generation pauses until the allowance renews.
- Trial allowance used up — the same: you keep chatting on Instant until the trial converts to your plan on day 7 and the full monthly allowance starts.
- Instant's daily fair use reached — rare in normal use; chat pauses until the next day.
Heavy use is also subject to our Fair usage policy.
After the reset
Premium models come back automatically when your daily cap or monthly allowance renews — your picked model is kept, so the next reply after the reset uses it again.