Transparency note: this site uses Claude, made by Anthropic, and the article was written with its help. So we won’t claim Claude is “the best for everyone”: the sources don’t support that. We’ll show where it stands out, with the source for each figure, and where ChatGPT or Gemini make more sense.
Updated: 30 October 2026 | CTRL + ALT News — Artificial Intelligence
If you can’t decide between Claude, ChatGPT and Gemini, the short answer is: it depends on what you do day to day. All three have become very good, and the choice is rarely between “good” and “bad”. It’s between tools that are stronger at different things.
This guide uses plain English and each company’s official pages, read on 30 October 2026, plus tests from an independent evaluator.
The quick answer
| If you want to… | Start with | Why (in short) |
|---|---|---|
| Code, or use a coding agent | Claude | Leads Artificial Analysis’s coding-agent index |
| Read and summarise very long documents without paying | Claude or Gemini | Anthropic lists a context window of up to 1 million tokens on every plan, and Gemini Pro also has 1 million |
| Create images and video | ChatGPT or Gemini | Both list image generation (and Gemini lists video); we didn’t find it on Claude’s page |
| Use AI inside Gmail and Google Docs | Gemini | It’s the one built into Google’s apps |
| Pay as little as possible | Gemini (AI Plus) or ChatGPT Go | The entry plans cost far less than Claude Pro |
| A general-purpose assistant with voice and many features | ChatGPT | It lists the most consumer features |
If none of these describes you, the section “How to choose in three questions” will help. And you can try all three for free.
First, the basics: what are Claude, ChatGPT and Gemini?
All three are AI assistants that chat with you, write, summarise, explain and help you code. Each is a product built on a language model (an LLM):
- Claude is made by Anthropic. Its current models are called Fable, Opus, Sonnet and Haiku.
- ChatGPT is made by OpenAI and uses the GPT family of models (GPT-6 and others).
- Gemini is made by Google, with models such as Gemini 4 Argon.
If those terms are new, see our explainer on LLMs, AI agents and agentic AI.
Plans and prices side by side
The figures below come from the official pages. They use different currencies, so check the price at the moment you subscribe.
| Claude | ChatGPT | Gemini | |
|---|---|---|---|
| Free | $0 (Sonnet and Haiku models) | £0 (“unlimited” text chats, subject to abuse guardrails) | £0 (Flash-Lite model) |
| Entry | Pro: $20 a month ($17 on the annual plan) | Go: £7 (may include ads) and Plus: £20 | AI Plus: £4.49 and AI Pro: £18.99 |
| Top tier | Max: from $100 | Pro: from £89 | Ultra: from £79.99 |
| Context window | Up to 1 million tokens, even on the free plan | 256,000 (Plus) and 400,000 (Pro) in reasoning mode | 1 million |
| Training on your data | You can opt out | You can opt out | We did not check |
What’s a “context window”? It’s how much text the model can hold in mind at once. A million tokens is about 555,000 words on Anthropic’s latest counter, or hundreds of pages. The bigger it is, the easier it is to paste in a whole book or contract.
Six practical situations: who does best at what
The examples below are illustrative. The conclusions come from what the sources say, and we say so when there is no data.
### 1. You code (or want to learn)
Where Claude stands out. In the evaluation aimed at coding agents, Artificial Analysis put Claude Code with Claude Opus 5.5 first, on 62 points, against 59 for Gemini 4 Carbon with Antigravity (comparison of 28 October). ChatGPT has Codex and Gemini has Antigravity, so both are also real options for coding. On Claude, Claude Code comes with the paid plans (Pro and Max).
Example: your website has a strange bug that spans several files. A coding agent reads the project, finds the problem, proposes a fix and runs the tests.
### 2. You need to read long documents
Claude and Gemini are ahead. Claude’s page lists a context window of up to 1 million tokens on every plan, including the free one. Gemini Pro also has 1 million. On ChatGPT, the fast-response mode has a much smaller window (27,000 tokens on the free plan, 54,000 on Plus), and reasoning mode reaches 256,000 on Plus.
Example: pasting in an 80-page contract and asking “which clauses work against me?”. The bigger the window, the less you have to split the text.
### 3. You write texts and emails
All three do well here, and it’s a matter of taste. We found no independent test comparing the writing quality of the three, so we won’t name a winner. The advice is simple: ask all three for the same text and pick the style you like best.
Example: drafting a polite reply to an annoyed customer. Try it on all three and see which sounds most like you.
### 4. You research and study
Gemini and ChatGPT offer deep research (Deep Research) on the free plan, with limits. On Claude, the research tool (Research) is on the paid plans. If you want a research report without paying, Gemini and ChatGPT have the edge.
Example: “compare the pros and cons of three types of travel insurance”. A deep-research feature searches several sources and returns a summary with references, which you should check.
### 5. You want to create images or video
ChatGPT and Gemini. ChatGPT lists image generation (limited on free, unlimited on Pro), and Gemini lists image generation and editing on the free plan and video on the paid plans. We did not find image generation on Claude’s page, which doesn’t prove it doesn’t exist, but it’s a signal for anyone who depends on it.
Example: creating an illustration for an invitation or a short video for social media.
### 6. You live in the Google ecosystem
Gemini. Gemini is built into Gmail and Docs (on the Pro plan) and includes cloud storage: 15 GB on free and 5 TB on Pro. If you already use Gmail, Docs and Drive every day, it’s the most natural route.
Example: asking for a summary of the latest email threads about a project without leaving Gmail.
What independent tests say
Artificial Analysis, an independent evaluator, publishes an overall index that combines nine evaluations. On 29 October the leaderboard showed:
| Model | Overall index v5 |
|---|---|
| Gemini 4 Carbon | 59 |
| Claude Opus 5.5 | 58 |
| GPT-6 Astra | 57 |
| GPT-6.1 Sol | 55 |
| Claude Sonnet 5.5 | 54 |
Two important cautions. First, Gemini 4 Carbon, which leads, has restricted access: Google released it only to a group of cyber defenders, so most people can’t use it today. Second, the numbers depend on each model’s settings and don’t measure the “experience” of using the product. Among the models the public can use, Claude Opus 5.5 appears just ahead.
That helps explain the title: Claude wins at coding with agents and sits near the top overall, but it doesn’t win at everything.
How to choose in three questions
- What will I mainly use it for? Coding or long documents: start with Claude. Images and video: ChatGPT or Gemini. Gmail and Docs: Gemini.
- How much can I spend? If not much, start with the free plans of all three and, if you need more, look at AI Plus (Gemini) and ChatGPT Go. Claude Pro costs $20 a month.
- Can I test first? Yes. All three have a free plan. Do the same real task on all three for a week and compare.
Can you use more than one?
Yes, and many people do: one assistant for coding, another for research and images. No rule says you must pick just one.
Frequently asked questions
### Is Claude better than ChatGPT?
It depends. At coding with agents, Claude is ahead on Artificial Analysis’s index. On consumer features such as image generation and voice, ChatGPT lists more. For writing, no independent test settles it.
### What’s the best free AI?
All three have a free plan, with different limits. Gemini includes image generation and deep research on the free plan, and Claude lists a context window of up to 1 million tokens even on free. Test all three on the same task.
### Does Claude generate images?
We did not find image generation on Claude’s official page. If you need it, check in the product itself before subscribing.
### Is Claude Pro worth paying for?
If you code often, use long documents or want Claude Code and Claude in Chrome, which come with the paid plans, it may be. If you use it only occasionally, the free plan may be enough.
### Is my data used to train the AI?
According to the official pages, Claude and ChatGPT let you opt out of your conversations being used for training. Read each company’s policy before sending sensitive data.
### Will the ranking change?
Very likely. Models and plans change within weeks. That’s why we give the research date and the sources.
What this means
Our reading: Claude is a strong choice for people who code and for people who work with long texts, and independent data supports that. But “the best” depends on what you do: anyone who needs images, video or the Google ecosystem has reasons to prefer ChatGPT or Gemini. The best test is your own: use the same real task on all three.
What we could not confirm
Plans and prices come from official pages that carry no date, read on 30 October 2026, in different currencies. The Gemini page was read in part. We did not test the three side by side, found no independent test of writing quality and did not check Gemini’s use of data for training. We did not find image generation on Claude’s page. Artificial Analysis results depend on each model’s settings.
### Sources and further reading
Leave a Reply