General assistant

How to use
ChatGPT.

Broad everyday work: drafting, file analysis, research, image work, voice, and structured problem-solving.

What it isgeneral AI assistant Workflows2 UpdatedJuly 2026

Inside the product

See the working surface.

A current first-party interface reference, preserved locally with its source and capture date.

Three official ChatGPT iPhone screens showing image creation, visual learning, and a structured training answer.

ChatGPT on iPhone, showing three common ways to move from a prompt to a useful visual or structured result.

Interface reference · Official ChatGPT App Store listing ↗ · Captured August 2, 2026

ChatGPT is a general AI assistant. Broad everyday work: drafting, file analysis, research, image work, voice, and structured problem-solving. This guide covers the whole path in one place: official access, a first session that produces something reviewable, the checks that make output trustworthy, and the permissions worth limiting before you connect real work.

Troiana principle

AI should make the work easier to inspect. If the workflow removes the source, the owner, or the review step, redesign the workflow.

01

Access & setup

Find, install, and sign in to ChatGPT

Get into the official ChatGPT experience with the right account and a setup you understand.

  1. 01

    Start at https://chatgpt.com/ and confirm the domain before entering account or payment information.

  2. 02

    Availability: Windows and macOS.

  3. 03

    OpenAI documents macOS 14 or later for the current Mac app. The download page selects the installer for Apple Silicon or Intel; Windows requirements and feature availability vary by release and plan.

  4. 04

    Sign in with the account you intend to keep using, then review plan, data, notification, and permission settings.

  5. 05

    Run one low-risk test task before connecting sensitive files, repositories, or workspace data.

  • Use official download pages.
  • Review permissions during setup.
  • Keep installers and applications updated.
02

First session

Your first useful ChatGPT session

Learn the interaction loop using a small task with a clear outcome.

  1. 01

    Open a new chat and state the outcome, audience, source material, and constraints.

  2. 02

    Ask for a short plan before requesting the full deliverable.

  3. 03

    Attach the actual file or source material instead of summarising it from memory.

  4. 04

    Check factual claims and use the final response as a draft to review, not an authority.

  • State the outcome before the background.
  • Provide the real source material.
  • Review the result before expanding the task.
03

Quality control

Check the quality of ChatGPT output

Turn a plausible draft into work you are willing to stand behind.

  1. 01

    Restate the original outcome and acceptance criteria before reviewing the response.

  2. 02

    Mark every factual claim, calculation, quotation, or figure that could materially affect the result.

  3. 03

    Check those items against the original source, a primary reference, or direct inspection.

  4. 04

    Ask the tool to list its assumptions and uncertainty, but never rely on self-review as the only check.

  5. 05

    Complete a human edit for relevance, tone, privacy, accessibility, and the exact delivery format.

  • Verify consequential claims independently.
  • Fluency is not accuracy; confident phrasing is not evidence.
  • Keep a record of sources and review decisions.
04

Privacy & permissions

Use ChatGPT safely

Use the platform productively without giving it unnecessary access or authority.

  1. 01

    Classify the task and its data before uploading or connecting anything.

  2. 02

    Review the product, account, organisation, retention, and training settings that apply to you.

  3. 03

    Grant the narrowest file, workspace, microphone, screen, or integration permissions needed.

  4. 04

    Remove secrets and personal or regulated information unless your approved policy explicitly permits it.

  5. 05

    Review the output and revoke permissions or disconnect sources that are no longer needed.

  • Model and tool availability change by plan and region. A fluent answer can still be wrong, so verify consequential claims against primary sources.
  • Follow your organisation's approved-use policy.
  • Never treat fluent output as authorization to act.
05

Models & pricing

What ChatGPT costs, model by model.

OpenAI's current frontier family is GPT-5.6, released 9 July 2026 in three tiers — Sol, Terra, and Luna — which differ in capability, latency, and price rather than being separate model families.

GPT-5.6 family and current API models
ModelRoleInputOutputLong-context in/out
GPT-5.6 SolFlagship — hardest reasoning, coding, science$5.00$30.00$10.00 / $45.00
GPT-5.6 TerraBalanced default for production workloads$2.00$12.00$4.00 / $18.00
GPT-5.6 LunaHigh-volume, latency-sensitive, low cost$0.20$1.20$0.40 / $1.80
GPT-5.5Previous flagship$5.00$30.00
GPT-5.4Previous balanced tier$2.50$15.00
o3Reasoning model (post price cut)$2.00$8.00
GPT-4oLegacy multimodal$2.50$10.00
GPT-4o miniLegacy low-cost$0.15$0.60
Long-context rates apply above the standard context threshold. Cached input bills at 10% of the standard input rate, and the Batch API halves both input and output.
Which GPT-5.6 tier each ChatGPT plan reaches
PlanModel accessBest for
FreeLuna, limited TerraOccasional questions and drafting
PlusTerra, limited SolIndividual daily professional use
ProSol, including extended reasoningHeavy research, coding, and analysis
Business / EnterpriseSol and Terra with admin controlsTeams needing data controls and SSO
ChatGPT is the consumer and team product built on these same models. Plan availability changes frequently and model access is the main practical difference between tiers.

Figures verified 2 September 2026 against OpenAI API pricing ↗. Model pricing changes frequently — confirm before you budget.

Frontier and default-tier pricing across the three major providers
ProviderFrontier tierBalanced defaultLow-cost tier
OpenAIGPT-5.6 Sol — $5 / $30GPT-5.6 Terra — $2 / $12GPT-5.6 Luna — $0.20 / $1.20
AnthropicFable 5.1 — $10 / $50Sonnet 5 — $2 / $10Haiku 4.5 — $1 / $5
GoogleGemini 3.1 Pro — $2 / $12Gemini 3.1 Pro — $2 / $12Gemini 3.5 Flash-Lite — $0.15 / $1.25
Input / output USD per million tokens, standard context, before batch or caching discounts. Compare tiers by role rather than by name — vendors label their tiers differently.

Cross-provider comparison verified 2 September 2026. Sources: OpenAI API pricing ↗ · Claude platform pricing ↗ · Gemini API pricing ↗

06

Common questions

ChatGPT questions people actually ask.

What is the difference between GPT-5.6 Sol, Terra, and Luna?

They are three tiers of the same GPT-5.6 family, separated by capability and cost rather than by purpose. Sol is the flagship for the hardest reasoning, coding, and research work at $5 per million input tokens. Terra is the balanced default for production applications at $2. Luna is optimised for speed and volume at $0.20, for simple high-frequency tasks where cost per request dominates.

Which GPT-5.6 tier should I actually use?

Start on Terra. It is the intended default and handles most production work. Move to Sol only when you can demonstrate Terra failing a task you can measure, and move to Luna when the task is simple, repetitive, and high-volume. Choosing Sol by default is the most common way teams overspend on the OpenAI API.

How much does the OpenAI API cost per million tokens?

As of September 2026, from $0.20 per million input tokens on GPT-5.6 Luna to $5 on GPT-5.6 Sol, with output tokens billed at roughly six times the input rate. Cached input bills at 10% of standard, and the Batch API halves both directions.

How do I reduce OpenAI API costs without changing model?

Three levers, in order of impact: use the Batch API for anything not needed in real time to halve the bill; structure prompts so the stable portion is cached, billing at 10% of standard input; and cut output tokens, which cost about six times what input costs, by requesting tighter formats.

Is ChatGPT Pro worth it over Plus?

Only if you are regularly hitting the limits of the Terra tier on hard reasoning or long coding tasks. Pro's value is access to Sol and extended reasoning. If your work is drafting, summarising, and everyday questions, Plus reaches the same quality for a fraction of the cost.

Does ChatGPT train on my conversations?

It depends on the plan and your settings. Business and Enterprise plans do not train on your data by default; consumer plans have a data-controls setting that governs it. Verify the setting on the account you actually use before uploading client material, and never rely on a plan's marketing page as evidence.

07

Core workflows

Step-by-step ways to use ChatGPT for the work it does best.

Each workflow is a separate guide with its own steps and review checkpoints.

08

Official references

Check the current product documentation.

Features, plan limits, availability, and data controls change. These official pages are the starting points used for this guide.

Explore the Troiana AI Hub →