WritingmateWritingmate Documentation

Fair Use Policy

Writingmate is moving from daily message limits to a simple monthly credit system. New plans use credits. Some older plans still use daily limits until they renew. This page explains both.

Credits

Credits are like minutes on a phone plan. Small tasks use less. Big tasks like video use more. They refill on your billing date each month, and unused plan credits do not carry over.

Every plan comes with a monthly pool of credits. You can spend them on anything: chat, images, video, voice, or web search. There are no daily caps — just one pool that refills on your billing date.

Video comes out of the same pool as everything else — there is no separate monthly count limit on credit plans. You can have at most 3 videos generating at the same time, which is a queue limit rather than an allowance. Plans still on daily limits keep their old monthly video count until they renew.

For per-model rates, how many messages each plan buys, and why a long thread costs more than a short one, see Understanding Credits.

What things roughly cost

What a message costs depends on the whole conversation, not just what you type. Every reply re-reads the thread so far, so the first message in a chat is cheap and the twentieth is not. A typical turn re-reads about 7,500 tokens — around 5,500 words — and writes about 600 back.

  • A chat message: well under 1 credit on our everyday models (GPT-5.6 Luna, Qwen3.7 Flash, Gemini 3.5 Flash Lite) and about 3 to 4 on top models like Claude Opus 5 or GPT-5.6 Sol. The difference is mostly the conversation: every message re-reads the chat so far. We cache that history so you are not charged full price for re-reading it, which is why a long thread costs less per message than its length suggests.
  • An image: about 2–20 credits, depending on the model. (One model, FLUX Schnell, costs 1 — it is available through our API only, not the image picker.)
  • Video: about 4–50 credits per second, depending on the model.
  • A web search: about 2 credits.

Bigger models cost more credits than smaller ones. If you want your credits to last, pick a lighter model for simple tasks.

No daily limit on top models

Older plans capped the strongest models — like Claude Opus — at 5 messages a day, no matter how much you paid. That cap is gone. Every model in our catalog is included on every paid plan, and you can use any of them as much as you like until your credits run out. (A small number of models are not in the catalog at all — we leave out models priced above our ceiling, and a few we cannot resell.)

The trade is simple and worth knowing before you start: strong models cost more credits per message than light ones, and long chats with big files cost more than short ones. A day of heavy Opus work will use a lot more of your pool than a day of quick questions. Nothing stops you — you just see it in your balance.

When your credits refill

Your plan credits refill on your billing date every month. Unused plan credits do not carry over.

If you run out

You have three options:

  • Wait. Your credits refill on your billing date.
  • Add credits. Buy a top-up pack on the upgrade page. Credits you buy last 90 days from purchase, and they are used after your plan credits run out. Bonus credits we add — such as the extra month yearly plans start with — have their own expiry, which is 12 months for the yearly welcome bonus.
  • Move to a bigger plan with more monthly credits.

If your plan is already on credits, you can see your balance any time in Profile Settings → Subscription. Older plans show daily limits there instead until they renew onto credits.

Older plans (daily message limits)

If your plan still shows daily limits, this applies to you until your plan renews.

Unlimited* is intended for normal use. It does not mean that one account can send an unlimited amount of model input or avoid all usage safeguards. Writingmate applies a daily fair-use limit so unusually heavy use does not reduce service quality for other users.

Where Unlimited* applies

On current public plans, Unlimited* appears for:

  • Basic-model use on paid plans
  • Pro-model use on the Ultimate plan

Other model categories can have a visible daily allowance. Legacy, promotional, and AppSumo plans can use different allowances or monthly pools. Check Profile Settings → Subscription and the model usage indicator in chat for the terms that apply to your workspace.

Writingmate does not publish a separate numeric limit for usage marked Unlimited*. If the fair-use limit is reached, the product reports a daily message limit instead of showing the internal limit.

What uses the allowance

Writingmate converts text-model usage into counted messages. A counted message is a unit based on the number of tokens processed. A token is a small piece of text that an AI model reads or writes.

Each request includes:

  • your new message
  • relevant chat history
  • attached file content and other context
  • the model's response

One counted message covers up to 16,000 effective tokens. Longer requests use more than one counted message. Cached prompt tokens receive the discount described in Message Limits.

Writingmate chat and text requests made through the OpenAI-compatible API share the applicable model-category allowance.

This counting rule also applies when the plan shows Unlimited*. Normal conversations usually use the fair-use allowance more slowly than requests with large amounts of repeated context.

Why files and Projects use more

A file adds its extracted content to the model's context. Follow-up requests can include that file content again along with the growing chat history. Large documents can therefore make one request count as several messages.

A chat inside a Project receives the Project's available file content as context. If a Project contains many or large files, each request in that Project can process substantially more text than the same request in a chat without those files.

Attaching a file does not automatically end unlimited access. File size, the amount of extracted text, chat history, the model's response, and repeated requests all affect how quickly usage reaches the fair-use limit.

Use the allowance efficiently

  • Attach only the files and sections needed for the task.
  • Keep unrelated files in separate Projects.
  • Use a chat outside a Project when Project files are not needed.
  • Start a new chat when earlier history is no longer useful. This does not reset usage already recorded, but it reduces the context in later requests.
  • Use a Basic model when the task does not require a Pro or Ultimate model.

If you reach the fair-use limit

The chat shows Daily message limit reached for the affected model category. You can:

  • switch to another model category that still has an available allowance
  • wait for the daily allowance to reset at 00:00 UTC
  • move to a plan with a higher applicable allowance, if one is available

Starting a new chat does not restore an allowance that has already been used.

AppSumo plans use monthly message pools with daily pacing limits rather than the current public-plan allowances. See AppSumo Plan Tiers for those rules.

Why this policy exists

AI requests consume shared processing capacity. The fair-use limit allows Writingmate to offer normal use as Unlimited* while protecting service availability from unusually large or repeated workloads.

Related docs