Assistants & Chatbots

How to choose a chatbot

Choosing between ChatGPT, Claude and Gemini is mostly a question of what you need the plan to do and what you will pay for it, not of which one is "smartest". The guides in this section compare the plans the vendors publish: price, models, usage limits, context windows, coding tools and what each company says about your data. Every figure links to the vendor's own page and carries the date we checked it. We have not run a hands-on quality test, so we do not rank assistants by how well they write or reason; we rank nothing we cannot source.

Where to start

If you already know the assistant you are considering, start with its pricing guide:

1. ChatGPT pricing breaks down Free, Go, Plus, Pro, Business and Enterprise: what each costs from Spain and in the US, which models and context windows each plan includes, how "unlimited" chats are defined and how the euro prices work.
2. Claude pricing does the same for Claude's Free, Pro, Max, Team and Enterprise plans, explains the five-hour usage window and runs three sample workloads through our API cost calculator to show when the API is cheaper than a plan.

• Microsoft Copilot pricing untangles the free Copilot app, Microsoft 365 Personal, Family, Premium and Pro, and the business plans, including who in a family actually gets Copilot.
• Gemini pricing covers Free, Google's Plus plan, Pro and Ultra in the US and Spain, what each bundles, and how the compute-based limits and October's switch to Flash-Lite for free users work.

If you are weighing one against the other, go to the comparisons:

3. Claude vs ChatGPT lines up the two main paid plans, their context windows (up to 1M tokens for Claude against 27K to 400K for ChatGPT), their coding agents and their data settings.
4. ChatGPT vs Gemini compares OpenAI's plans with Google's Google's Plus plan, Pro and Ultra, which bundle the assistant with storage and Google apps, and explains how Google's compute-based limits differ from ChatGPT's "unlimited" text chats.
5. ChatGPT alternatives covers seven options, from Claude and Gemini to Microsoft Copilot, Perplexity, Mistral, DeepSeek and a local model, by free plan, price and what each does best.

• Perplexity vs ChatGPT compares an answer engine built around cited web search with a general assistant: free limits, Pro and Plus prices, citations and privacy.
• Claude vs Gemini closes the triangle with the two comparisons above: Claude Pro against Google's Pro plan, what each bundles (Claude Code, or storage and Gemini in Gmail and Docs), context windows and how each uses your chats for training.
• DeepSeek vs ChatGPT puts data first: DeepSeek stores personal data in China and Italy's regulator limited it in 2025, while ChatGPT lets you switch off training; plus plans, models and API prices from both companies' documents.
• Best free chatbot compares the free plans of ChatGPT, Gemini, Claude, Copilot, Mistral Vibe and DeepSeek: what each includes, how the limits work and how each uses your chats.
• Chatbots for Excel and Sheets compares Copilot in Excel, Gemini in Google Sheets, ChatGPT's data analysis and Claude for Excel: where each runs, which plan it needs and what it cannot do.
• Best Image Generators: ChatGPT vs Gemini vs Midjourney compares ChatGPT, Gemini, Midjourney and Adobe Firefly: plans, limits, usage rights and which one to pick.
• ChatGPT Memory: How It Works and How to Turn It Off explains saved memories and chat history in ChatGPT, how to see and delete them, and how Claude and Gemini handle memory.
• Make a Presentation with ChatGPT, Copilot or Gemini compares ChatGPT, Copilot in PowerPoint, Gemini in Google Slides and Claude for building slides, plan by plan.
• Video Generators: Gemini Omni, Runway, After Sora compares the video tools to use now that Sora is closed: Gemini Omni, Runway, Midjourney and Firefly, with their plans and credits.

What we compare, and what we leave out

Plans change every few months, so each guide separates what the vendor publishes from what we calculate. Prices are shown as the vendor's page shows them: ChatGPT and Claude euro prices from Spain, dollar prices for the US, and we say when a price does not include tax. OpenAI's pages do not say whether euro prices include VAT, while Anthropic says its prices exclude tax, so the number on the checkout screen is the one that counts. We include the limits that decide whether a plan works for you (messages, five-hour windows, context size) because the headline price rarely predicts that.

We leave out quality rankings, benchmark leaderboards and promotional offers. A benchmark score does not tell you whether an assistant will fit your work, and a promotion disappears. What we can document is what each plan includes and what it costs, and that is what you pay for.

Which questions these guides answer

• Is the free plan enough? Both ChatGPT and Claude have a free plan, and the pricing guides list what it includes and what it limits. A week of real use on your own tasks is a better test than any table.
• What do I get for the next price step? The guides show what changes between a low-cost plan, the main paid plan and the heavy-use tiers, so you can see whether a higher price buys more usage, better models or extra tools.
• Which one fits my documents? Context windows differ by plan and by company, and they decide how much text you can paste in at once. For the concept, see what a context window is.
• What about my data? Each guide records whether an opt-out from model training exists. For ChatGPT's full data settings and your rights in Europe, see is ChatGPT safe.
• Subscription or API? A subscription has usage limits and a flat price; an API bills every token. The LLM API pricing guide and its calculator help you compare.

What to read next

If you want the assistant to write code, read best coding assistant. If you would rather run a model on your own computer, start with how to run an LLM locally. And if you are choosing for a company, check the rules that already apply to businesses that use chatbots in our Regulation (EU) 2024/1689 guide.