NEW · Morning journal prompts → start your day with intention
Random Prompts
New default: GPT-5.5 Instant replaced GPT-4o on May 5, 2026

GPT-5.5 Prompt Generator

The GPT-5.5 prompt generator gives you 20 free, copy-ready prompts for OpenAI's new default ChatGPT model. Creative writing, business docs, coding, research, and marketing — use them directly in ChatGPT, no signup required here.

What is the GPT-5.5 Prompt Generator?

The GPT-5.5 prompt generator on this page provides 20 professionally structured prompts for OpenAI's latest default model — GPT-5.5 Instant, which became the default ChatGPT model on May 5, 2026. GPT-5.5 is a hybrid reasoning and generation model: it inherits the deep multi-step reasoning of GPT-5 but delivers outputs at a speed and fluency closer to GPT-4o, making it the most capable everyday AI model OpenAI has shipped.

Compared to GPT-4o, GPT-5.5 follows longer and more nuanced instructions without losing context, produces more natural prose that avoids AI-writing clichés, and handles complex structured outputs (pitch decks, legal summaries, technical documentation) with noticeably less hallucination. OpenAI also published a dedicated GPT-5.5 prompting guide noting that several prompting tactics that worked on older models actively backfire on GPT-5.5 — particularly over-simplifying your instructions or using keyword-stuffed prompts.

Every prompt below is copy-ready and structured to take full advantage of GPT-5.5's capabilities: clear audience definition, explicit format constraints, tone instructions, and a specific output length. Paste directly into ChatGPT at chat.openai.com — GPT-5.5 Instant is the default, so no model switching needed.

How to Write a GPT-5.5 Prompt

GPT-5.5 is specifically tuned to follow detailed, structured instructions. The more precisely you define the output, the better it performs. Use this framework:

[Task verb] + [specific subject/context] + [audience definition] + [format constraints] + [tone/style] + [length target] + [what NOT to do].

What GPT-5.5 Handles Better Than GPT-4o:

  • Multi-part structured documents (10+ sections)
  • Sustained narrative voice at 1,000+ words
  • Complex conditional instructions ("if X, do Y")
  • Precise format outputs (tables, JSON, scripts)
  • Counter-intuitive or anti-cliché creative briefs
  • Fact-checking its own claims mid-generation

GPT-5.5 Prompting Tips:

  • Define the target reader explicitly (not just "a blog post")
  • Specify word count — GPT-5.5 respects it more accurately
  • State what to avoid (clichés, passive voice, jargon)
  • Use "Write" not "Can you write" — direct imperatives work better
  • Give a style reference: "in the style of [publication/person]"
  • Break multi-task prompts into numbered steps

Key difference from GPT-4o prompting:

GPT-4o responded well to short, vague prompts and filled in the gaps itself. GPT-5.5 rewards specificity: the more constraints and context you provide, the more it outperforms GPT-4o. Vague prompts still work, but you're leaving most of GPT-5.5's capability on the table.

20 Free GPT-5.5 Prompts — Copy & Paste

Click any prompt to copy — paste directly into ChatGPT (GPT-5.5 is now the default model)

1. Long-Form Short Story

Creative Writing

Write a 1,200-word short story about a deep-sea marine biologist who discovers a bioluminescent organism that responds to human music. The story should open in medias res during a night dive, use close third-person POV, and build to a moment of quiet wonder rather than a dramatic climax. Tone: lyrical, grounded in real marine science, no fantastical elements beyond the organism itself. End with an image, not a resolution.

2. Series Bible — TV Pilot

Screenwriting

Write a 600-word series bible for a one-hour prestige drama set inside a modern algorithmic trading firm during the 72-hour window of a flash crash. The main character is a mid-level quant analyst who suspects the crash is not accidental. Include: logline, tone paragraph, three main characters with 2-sentence bios, five season-arc story questions, and a network pitch comparison ('it's X meets Y for Z audience').

3. Venture Capital Pitch Deck Narrative

Business Writing

Write the narrative copy for a 10-slide venture capital pitch deck for a B2B SaaS company that automates compliance documentation for mid-market healthcare providers. The company has $2.1M ARR, 40% YoY growth, and is raising a $12M Series A. Each slide gets a headline (max 8 words) and 3–4 bullet points (max 12 words each). Slides: Problem, Solution, Market Size, Product Demo, Business Model, Traction, Competition, Team, Use of Funds, Ask.

4. Technical API Documentation

Technical Writing

Write production-ready documentation for a REST API endpoint: POST /v1/payments/initiate. Include: endpoint description (2 sentences), authentication requirements, request body schema (JSON, with field names, types, required/optional flags, and validation rules), response schema for 200 OK and error codes 400/401/422/500, a cURL example request, a JavaScript fetch example, and a 'Common Mistakes' section with three items. Follow Stripe's documentation style.

5. Cold Email Sequence (3-Email)

Marketing Copy

Write a 3-email cold outreach sequence for a SaaS tool that helps e-commerce brands reduce return rates using predictive sizing AI. Target persona: VP of Operations at a DTC fashion brand doing $5M–$50M revenue. Email 1 (day 1): problem-led, under 80 words, single CTA to book a call. Email 2 (day 4): social proof / case study hook, under 100 words, same CTA. Email 3 (day 9): breakup email with a light touch, under 60 words. Subject lines included. No jargon, no buzzwords.

6. Code Review with Explanation

Coding

Review the following Python function for correctness, performance, and edge cases. For each issue found: (1) describe the problem in plain English, (2) explain why it matters, (3) provide the corrected code. After the review, rewrite the entire function with all fixes applied and add inline comments only where the logic is non-obvious. Use PEP 8 style. [PASTE YOUR FUNCTION HERE]

7. Research Synthesis Memo

Analysis

Synthesize the following 5 research abstracts into a 500-word executive memo suitable for a non-technical C-suite audience. Structure: (1) one-paragraph 'What We Know' summary, (2) a 'Key Disagreements' section noting where findings conflict, (3) a 'Confidence Level' assessment (high / medium / low) with justification, (4) two 'Questions for Further Research.' Avoid passive voice. Do not use the word 'utilize.' [PASTE ABSTRACTS HERE]

8. Job Interview Preparation

Career

I am interviewing for a Senior Product Manager role at a Series B fintech startup focused on SMB lending. The job description emphasizes: 0→1 product experience, data-driven decision-making, and stakeholder alignment with engineering and risk teams. My background: 4 years PM at a consumer app, no fintech experience. Give me: (1) the 5 most likely behavioral questions for this role with STAR-format answer frameworks tailored to my background, (2) 3 technical/domain questions I should prepare for, and (3) 5 smart questions to ask them that signal strategic thinking.

9. SEO Blog Post — Full Draft

Content Marketing

Write a 1,400-word SEO blog post targeting the exact-match keyword 'AI tools for small business 2026'. Structure: H1 containing the keyword, a 150-word intro that opens with a statistic or provocative question (not 'In today's fast-paced world'), 6 H2 sections covering specific tools by category, a comparison table with 5 tools (columns: Tool, Best For, Free Tier, Starting Price, Standout Feature), and a 100-word conclusion with a CTA. No keyword stuffing. Flesch reading ease above 60.

10. Legal Document Plain-Language Summary

Legal

Summarize the following SaaS Master Service Agreement in plain English for a non-lawyer buyer. For each section: (1) a one-sentence summary of what it says, (2) a 'Watch Out' flag if it contains unusual or risky terms, and (3) a 'Questions to Ask' item where the buyer should push back. Organize by MSA section in the same order as the original. Use a table format: Column 1: Section Name, Column 2: Plain Summary, Column 3: Watch Out, Column 4: Questions to Ask. [PASTE MSA HERE]

11. Product Page Copy

E-commerce

Write product page copy for a $185 premium ceramic travel mug. Target customer: urban professionals who commute by train, care about sustainability, and buy from DTC brands. Include: (1) a headline under 10 words that leads with the emotional benefit (not the product name), (2) a 60-word subheadline expanding on the benefit, (3) a feature list of 6 bullet points in 'Feature — what it means for you' format, (4) a 120-word story paragraph written in second person, and (5) three social proof microcopy lines (under 15 words each) for use near the Add to Cart button.

12. Podcast Episode Script

Audio/Video

Write a 10-minute podcast script (approximately 1,400 words at 140 wpm) for a solo episode titled 'Why Most Goal-Setting Advice Fails (And What Actually Works)'. Structure: cold open with a counter-intuitive hook (no music cues needed), a 3-minute context section citing real behavioral science research, a 5-minute main framework section with 3 named principles, a 1-minute actionable takeaway, and a 30-second outro with a listener challenge. Conversational tone, no bullet-reading cadence. Use short sentences. Write it exactly as it would be spoken.

13. 30-Day Social Media Content Calendar

Social Media

Create a 30-day Instagram and LinkedIn content calendar for a personal finance educator targeting millennials earning $80K–$150K. One post per day. For each post include: (1) platform (Instagram or LinkedIn), (2) content format (carousel, single image, text-only, Reel hook), (3) topic, (4) opening line (the first sentence visible before 'more'), and (5) CTA. Organize in a table. Themes to cover across the month: investing basics, debt paydown, tax optimization, money psychology, and income growth. No repetitive CTAs.

14. Business Plan — Executive Summary

Business Writing

Write a 500-word executive summary for a business plan for a mobile dog grooming van service in Austin, Texas. The business targets working professionals with dogs in the suburbs and charges a $30 premium over salon prices for the convenience of at-home service. Include: the problem statement, solution, target market size (sourced reasoning acceptable), revenue model, competitive advantage, year-1 revenue projection with basic assumptions, and funding ask ($75K). Use confident, direct language. No passive voice.

15. YouTube Video Script — Hook + Outline

Video Content

Write the hook (first 90 seconds, approximately 200 words) and a full scene-by-scene outline for a 12-minute YouTube video titled 'I Tested 7 AI Writing Tools So You Don't Have To'. The hook should open with a pattern interrupt (not a question), create curiosity without clickbait, and set up a clear narrative promise. The outline should cover 7 tools with a consistent structure for each: name, test task, result, verdict (keep / skip). Include a B-roll note for each section. Tone: skeptical but fair, like a smart friend's honest review.

16. Customer Support Email Templates

Business Operations

Write 5 customer support email templates for a DTC furniture brand. Templates needed: (1) Order delayed beyond estimated date, (2) Item arrived damaged, (3) Return request approved, (4) Return request denied (outside return window), (5) Refund issued confirmation. Each template: subject line, greeting, body (under 120 words), closing, and a PS line that adds warmth. Tone: human, empathetic, no corporate jargon. Leave placeholders in [BRACKETS] for order numbers, names, and dates. Do not use 'We apologize for any inconvenience.'

17. Data Analysis Narrative

Analytics

I have the following dataset summary from a 6-month A/B test on our checkout flow: [PASTE DATA]. Write a 400-word narrative analysis suitable for presenting to non-technical stakeholders. Include: (1) a one-sentence headline result, (2) what the data shows (avoid 'the data shows'), (3) statistical significance note in plain English, (4) one important caveat about data limitations, and (5) a recommended next action. Use active voice throughout. Structure as flowing paragraphs, not bullet points.

18. Salary Negotiation Script

Career

Write a word-for-word script for a salary negotiation phone call. Scenario: I have a written offer for $105K. My research shows market rate is $118K–$128K. I have a competing offer I can reference at $112K. I want to reach $120K base + a $10K signing bonus. The script should cover: the opener (how to start the call), the ask (specific numbers, not ranges), handling the 'this is our best offer' objection, handling silence, and how to close whether they say yes, no, or propose a middle ground. Include notes on tone and delivery in [brackets].

19. Newsletter Edition — Full Draft

Email Marketing

Write a full edition of a weekly newsletter called 'The Contrarian' — an investing and personal finance newsletter for readers who are skeptical of conventional advice. This edition's theme: 'Why index funds might not be enough for people under 40.' Length: 700 words. Structure: subject line (A/B test two options), preview text (under 90 characters), a 200-word opening essay, a 'This Week's Contrarian Take' section with one argument and one counterargument, three curated links with 2-sentence summaries, and a one-paragraph closer. Tone: confident, evidence-based, no panic, no hype.

20. Academic Essay Outline

Education

Create a detailed essay outline for a 3,500-word undergraduate philosophy essay arguing that consciousness cannot be fully explained by computational processes (arguing against strong AI / functionalism). Include: thesis statement, five body sections with main claim, 2–3 supporting arguments, and 1 counterargument + rebuttal for each section, key philosophers and texts to reference per section (e.g., Searle, Nagel, Chalmers, Dennett), and a conclusion strategy. The outline should be detailed enough that writing the essay requires no further research planning.

GPT-5.5 vs. Other AI Models (May 2026)

Understanding where GPT-5.5 fits against current frontier models helps you choose the right tool for each task:

Model Best For Speed Context
GPT-5.5 Instant ★ Long-form writing, structured docs, everyday reasoning Fast 128K tokens
GPT-5 (full) Expert-level research, complex multi-step agentic tasks Slow 256K tokens
Claude Opus 4.7 Code, analysis, long document synthesis Medium 200K tokens
Gemini 2.5 Pro Multimodal (image/video input), Google Workspace tasks Fast 1M tokens
DeepSeek V4 Pro Code generation, open-weights local deployment Fast 128K tokens

★ GPT-5.5 Instant is the default ChatGPT model as of May 5, 2026. Available to all users including free tier.

GPT-5.5 Tips for Better Results

Do This:

  • Give the model a role: "You are a senior copywriter at a DTC brand"
  • Define the reader: "written for non-technical VPs of Operations"
  • Specify exact word or sentence counts
  • Tell it what to avoid ("no passive voice", "no bullet points")
  • Ask for A/B versions: "give me two versions, one formal, one casual"
  • Use follow-up prompts: "make it 20% shorter" or "punch up the opening"

Avoid This:

  • Over-simplifying: "write me a blog post about X" (use the full templates above)
  • Asking GPT-5.5 to be "creative" with no constraints — it interprets creativity broadly
  • Keyword-stuffed prompts that worked on GPT-3/4 — GPT-5.5 handles natural language better
  • Prompts that say "make it sound human" — it already does; specify the tone instead
  • Re-generating the same prompt hoping for a different result — revise the constraints

Frequently Asked Questions — GPT-5.5

What is GPT-5.5?

GPT-5.5 (also called GPT-5.5 Instant) is OpenAI's hybrid reasoning and generation model, released in April 2026 and made the default ChatGPT model on May 5, 2026. It combines GPT-5's deep reasoning capabilities with faster, more fluid text generation, making it significantly better than GPT-4o at complex writing, nuanced analysis, multi-step reasoning, and long-form content — while being more practical than the full GPT-5 for everyday tasks.

How is GPT-5.5 different from GPT-4o and GPT-5?

GPT-5.5 sits between GPT-4o and GPT-5. GPT-4o is fast and capable for everyday tasks but struggles with complex multi-step reasoning. GPT-5 is the most powerful but slower and more expensive. GPT-5.5 Instant delivers near-GPT-5 quality reasoning at a speed closer to GPT-4o — making it the best default model for most users. For prompt writing: GPT-5.5 handles longer, more nuanced instructions better than GPT-4o, and produces more natural, less robotic outputs than earlier models.

What kinds of prompts work best for GPT-5.5?

GPT-5.5 excels at structured, detailed prompts that give it clear constraints and a defined output format. It handles long-form instructions without losing context, follows complex multi-part briefs accurately, and produces coherent outputs at 1,000+ words without drifting. Unlike earlier models, GPT-5.5 benefits from prompts that specify the audience, tone, and format explicitly — not just the topic. The more context and structure you give, the more precise the output.

How do I get access to GPT-5.5?

GPT-5.5 Instant is available to all ChatGPT users including the free tier as the default model as of May 5, 2026. Full GPT-5.5 (with extended context and deeper reasoning) is available on ChatGPT Plus, Pro, and Team plans via the model selector. API access is available through the OpenAI API with the model ID 'gpt-5.5-instant' or 'gpt-5.5'.

What is the context window for GPT-5.5?

GPT-5.5 Instant supports a 128K token context window — large enough for a full novel or extensive document. The full GPT-5.5 model supports 256K tokens. This means you can paste in lengthy documents, entire codebases, or research papers and have GPT-5.5 analyze, summarize, or work with all of it in a single session.

Are GPT-5.5 outputs better than GPT-4o for creative writing?

Yes — significantly. GPT-5.5 produces more varied sentence structures, catches and avoids AI-writing clichés that plagued earlier models, and sustains narrative voice across long documents. For creative writing specifically, GPT-5.5's improved instruction-following means prompts that define tone, POV, and style constraints are executed more faithfully. The model also does a better job of subverting your expectations if you ask it to — older models defaulted to generic 'safe' outputs even when prompted for something unusual.

More AI Prompt Tools