control ChatGPT response length

Control ChatGPT response length get a paragraph, not an essay

Say the size you want and the model gives it to you. Stop skimming walls of text.

In short

To control ChatGPT response length, state the size in your prompt: 'in 100 words,' 'in three bullets,' 'one short paragraph.'

  • Be specific. A number is a rule; 'be brief' is a suggestion the model half-follows.
  • Ask for a bounded format, like 'a five-row table,' to cap size and shape at once.
  • If an answer gets cut off, type 'continue' and it picks up where it stopped.
The short version

Say the size you want

You control ChatGPT response length the same way you control everything else about the output: by asking. 'Explain this in 100 words,' 'give me three bullets,' 'one sentence.' Left unbounded, the model defaults to thorough, which reads as bloated when you wanted the short version. Naming the length up front is faster than trimming a wall of text after it lands. Length is one of several AI prompt constraints worth making standing rules, and the standing-rules slot is your ChatGPT system prompt.

The workflow

How to control ChatGPT response length

Give the model a number, a format, or a level of detail, and it holds the line.

  1. 1

    Name a word or item count

    'In 50 words,' 'no more than five bullets,' 'a single paragraph.' The number does the work a vague 'keep it short' cannot.

  2. 2

    Ask for a bounded format

    'A three-row table,' 'a numbered list of five,' 'a one-line summary.' A format caps the length and the shape together.

  3. 3

    Set the level of detail

    'Give me the headline only,' or 'a two-sentence overview, then detail if I ask.' Layering keeps the first reply short.

  4. 4

    Type 'continue' if it cuts off

    When a long answer stops partway, typing 'continue' resumes it from exactly where it left off.

  5. 5

    Save the rule if you always want short

    Put 'keep answers under 150 words unless I ask for more' in your Custom Instructions so every reply stays tight.

Watch-outs

  • 'Be brief' is weak. A number, like 'under 80 words,' is a rule the model actually follows.
  • Short does not mean vague. Pair a length cap with 'keep the specifics' so it trims fluff, not facts.
  • For a long deliverable, ask for an outline first, then expand only the parts you want.
FAQ

Common questions

How do I control ChatGPT's response length?

State the size in your prompt: a word count ('in 100 words'), an item count ('three bullets'), or a bounded format ('a five-row table'). The more specific the number, the more reliably the model holds to it.

Why are ChatGPT's answers so long?

Because with no length given, the model defaults to thorough and over-explains, so you control ChatGPT response length by saying it up front. It is not a bug so much as a missing instruction: tell it the size you want and the padding disappears.

How do I stop ChatGPT from cutting off mid-answer?

If a long response stops partway, type 'continue' and it resumes from where it left off. To avoid it entirely, ask for a shorter format or an outline first, then expand the sections you want.

Can I set a default response length?

Yes. Add a rule like 'keep answers under 150 words unless I ask for detail' to your Custom Instructions, and it applies to every chat automatically.

See how you actually work with AI

The AI Skills Quiz scores your real habits, including how precisely you steer the model's output. Free, and it takes a couple of minutes.

Professionals from these companies build their AI skills with Candova

Study.comMoneyGeekHatchOsteriaGuild CraftLearn.orgFGAParallelVestaboardRieckenProsperityStudy.comMoneyGeekHatchOsteriaGuild CraftLearn.orgFGAParallelVestaboardRieckenProsperity

The future belongs to AI‑native companies.
Make sure yours is one.

We make your people AI-fluent so your company becomes AI-native, with Cando, a personal AI trainer for every employee. Start with Candova AI today.

Free to start · Productivity gain guarantee · Built for your real job

Future-proof your work with AI

On-demand · start today

Get started