This free ChatGPT prompt generator writes prompts tuned to the model you use: GPT-6 Astra, GPT-6.1 Sol, GPT-6 Luna or a GPT-5 model. Pick the model and the task, describe what you need, and copy a prompt that follows OpenAI's own guidance for that model.
Models last verified:
Generate your prompt
Access 40,000+ AI Prompts for ChatGPT, Claude & Codex — Free!
Subscribe to get instant access to our complete Notion Prompt Library — the largest curated collection of prompts for ChatGPT, Claude, OpenAI Codex, and other leading AI models. Optimized for real-world workflows across coding, research, content creation, and business.
Access Free Prompt LibraryWhich GPT model should I choose?
Compare the GPT-6 and GPT-5 models covered by this generator by price, context window, reasoning effort and recommended use.
| Model | Price per 1M tokens (API) | Context window | Best use |
|---|---|---|---|
| GPT-6 Astra | $10.00 in / $50.00 out | 1,050,000 tokens | Complex reasoning, coding, computer use, research, and document creation. |
| GPT-6.1 Sol | $2.00 in / $10.00 out | 1,050,000 tokens | Complex coding, computer use, and professional work at lower cost than Astra. |
| GPT-6 Sol | $2.00 in / $10.00 out | 1,050,000 tokens | Complex coding and agentic workflows; OpenAI directs new Sol users to GPT-6.1 Sol. |
| GPT-6 Luna | $0.10 in / $0.50 out | 1,050,000 tokens | Focused, high-volume tasks; extraction, classification, and structured summaries at scale. |
| ChatGPT | Included in ChatGPT plans | Set by your ChatGPT plan | Prompts you paste into the ChatGPT app |
| GPT-5.6 Sol | $4.00 in / $20.00 out | 1,050,000 tokens | Flagship GPT-5.6 capability for complex professional work. |
| GPT-5.6 Terra | $2.00 in / $12.00 out | 1,050,000 tokens | GPT-5.6 workloads that balance intelligence and cost. |
| GPT-5.6 Luna | $0.20 in / $1.20 out | 1,050,000 tokens | Cost-sensitive, high-volume workloads. |
| GPT-5.5 | $5.00 in / $30.00 out | 1,050,000 tokens | Complex professional work. |
| GPT-5.4 | $2.50 in / $15.00 out | 1,050,000 tokens | Complex professional work. |
| GPT-5.4 Mini | $0.75 in / $4.50 out | 400,000 tokens | High-volume coding, computer use, and subagent work. |
| GPT-5.2 | $1.75 in / $14.00 out | 400,000 tokens | Previous flagship model for complex professional work; OpenAI recommends the latest GPT-6 Astra. |
How do I use my prompt?
- Choose ChatGPT or the GPT model you plan to use.
- Select a task type, difficulty and length.
- Describe your task and add any context, then generate your prompt.
- Copy the result and paste it into ChatGPT, or use the generated request with the OpenAI API.
Prompting tips for GPT-6 and GPT-5 models
- Use the instructions parameter or higher-authority message roles for the model’s behavioral goals; keep user input in input so instruction priority stays clear. Prompt engineering
- Organize developer prompts into Identity, Instructions, optional Examples, and Context. Markdown headings and XML-style tags help separate logical sections and supplied documents. Prompt engineering
- State the desired writing style and response structure, because GPT-6 Astra otherwise tends to produce detailed, highly formatted answers. Using GPT-6
- Describe the goal, relevant context, constraints, and what done looks like. Define approval boundaries rather than relying on blanket requests for permission. A practical guide to building with GPT-6
- Put stable instructions and reference material before changing task details to improve cache reuse; include cache-write and long-context costs in estimates. A practical guide to building with GPT-6
- Match reasoning effort to the job: low for routine work, medium for judgment, high for difficult analysis, and test xhigh or max only when high falls short. A practical guide to building with GPT-6
- Use model snapshots and representative evaluations when changing prompts or models so behavior changes are measurable rather than assumed. Prompt engineering
Worked examples
ChatGPT pricing brief
Before: Write a pricing brief for my meal-kit start-up.
After (made with this generator):
Task: Write a one-page pricing brief for [COMPANY], a meal-kit start-up launching in [MARKET]. Context: [Paste product details, costs and competitor prices.] Output: Lead with the recommended price. Then give three pricing options in a table (price, margin, who it suits), followed by two short paragraphs on risks. Use plain language, and use lists only for parallel items. Constraints: Use only the figures I provide. Label any assumption as an assumption. Done when: the brief fits on one page and every number traces back to my figures. In ChatGPT, choose a Thinking level, because this task needs judgment.
What the generator added: The prompt adds a one-page limit, a recommended price first, a three-option pricing table and two risk paragraphs. It restricts figures to those supplied, requires labelled assumptions and adds a done-when check that every number traces back to your figures.
Data Analysis
Before: Find the three most important sales trends this quarter.
After (made with this generator):
# Identity You are a senior data analyst for [COMPANY]. # Instructions - Lead with the main finding, then the evidence that supports it. - Default to clear, concise paragraphs. Use a table only to compare figures. - Do not use these words or framings: delve, foster, leverage, "it's worth noting", "In short". - Complete the analysis before asking any clarifying question. - Done when: every trend you report names the columns and rows it comes from. # Context <data> [PASTE THE TABLE OR CSV] </data> # Request Find the three most important sales trends this quarter. Use GPT-6.1 Sol at medium effort.
What the generator added: The prompt adds a senior analyst role, tagged data context and instructions to lead with the main finding and supporting evidence. It requires each trend to identify its source columns and rows, and the analysis to be completed before clarifying questions.
Customer Email Triage
Before: Choose a department for this customer email and check whether it is urgent.
After (made with this generator):
{
"model": "gpt-6-luna",
"input": "[CUSTOMER MESSAGE]",
"questions": [
{
"type": "choice",
"name": "department",
"instructions": "Which team should handle this message? Judge only by what the customer asks for.",
"choices": [
{
"value": "billing",
"description": "Payments, refunds, invoices and double charges"
},
{
"value": "technical",
"description": "Log-in problems, errors and bugs"
},
{
"value": "sales",
"description": "Pricing questions and new orders"
},
{
"value": "other",
"description": "Anything that fits none of the above; sent for human review"
}
]
},
{
"type": "predicate",
"name": "urgent",
"instructions": "Does the customer state a deadline today, money lost or a service outage? Ignore polite urgency such as 'as soon as you can'."
}
]
}
GPT-6 Luna, Decisions API (beta): POST /v1/decisions. $0.10 per 1M input tokens, no output charge.
What the generator added: This Decisions API request (beta, GPT-6 Luna) adds department choices with criteria for billing, technical and sales messages, plus an "other" choice sent for human review. It also asks a separate urgency question, answered as a probability, that ignores polite urgency.
Frequently asked questions
Useful links
- ChatGPT Image Prompt Generator
- Codex Prompt Generator and AGENTS.md Builder
- OpenAI Voice Agent Prompt Generator
- gpt-oss Prompt Generator
- Gemini Prompts Generator
- Midjourney Prompts Generator
- Free ChatGPT AI tools (all tools)
- Choose GPT-6 Sol, Luna, or Astra for ChatGPT Work and Codex: Capability, Cost, Reasoning Effort, Permissions, and Review
- OpenAI Introduces GPT-6.1 Sol for ChatGPT Work, Codex, and the API: Availability, Pricing, and Limits
- How to Build a ChatGPT Prompt Engineering System for Your Business: Complete ROI-Driven Framework