Ask ten people at any company which AI assistant they use and you will probably hear two names more than any others. Claude and ChatGPT. Both have become part of everyday work for writing and research and coding and a growing list of tasks that used to eat up someone's afternoon.
The comparison is also moving fast. Both companies have released major updates in just the last few weeks. So instead of declaring a winner this post looks at where each one stands right now and where each one genuinely fits.
Where Things Stand Today
OpenAI launched its GPT 6 family this month. The flagship is called Astra and it arrived in early September. Two lighter versions called Sol and Luna followed on September 22. In ChatGPT the top model is available on higher plans while Plus users reach it inside the Work and Codex modes of the desktop app. Free users get a lighter model with an option to think longer on harder questions.
Anthropic has been just as busy. Claude Opus 5.5 was released on September 22 and Anthropic says it matches the performance of its larger Fable model on most work while costing noticeably less to run. Claude Sonnet 5 remains the balanced everyday option and Haiku 4.5 covers fast and lightweight tasks.
On public benchmarks the two are close. One independent tracker this month put ChatGPT slightly ahead overall but noted the margin was narrow enough that it should not be treated as settled. Claude scored higher on some coding tests. ChatGPT scored higher on some agentic browsing tests. Benchmarks are useful but they rarely predict how a tool feels on your own real work.
Where ChatGPT Tends to Win
ChatGPT is the broader product. It handles images and voice and web search and file analysis in one place. If you need to generate images it is currently the only one of the two that can. Its image tools were updated again in early September. Voice conversations are also more mature. OpenAI has also shut down its video tools this year so neither product should be chosen for video.
It also has a large ecosystem of plugins and custom assistants and integrations. For a team that wants one app to do a bit of everything ChatGPT is the easier all purpose pick.
Where Claude Tends to Win
Claude is the more focused product and many reviewers say it shows in the quality of the work. Writers often report that its drafts need less editing and follow style instructions more closely. It tends to hold up well on long documents and multi step tasks where a lot of context has to stay in view.
Its agentic tools are a real strength too. Claude Code helps developers work directly on their projects and Claude Cowork brings similar delegation to people who do not write code. Claude also connects to outside tools through custom connectors and offers a memory import flow that makes switching from another assistant easier.
What About Price
For individuals this is nearly a tie. Claude Pro and ChatGPT Plus both cost about 20 dollars a month. Both companies also offer higher tiers at 100 and 200 dollars for heavy users. ChatGPT has a cheaper Go plan and a somewhat more generous free tier for casual use. Model access and usage limits shift during rollouts so it is worth checking the model picker before you subscribe for one specific model.
How a Business Should Actually Choose
The most useful question is not which tool is better. It is which tool fits the work your team does every day.
If your team writes a lot of reports and proposals and client documents or reviews long files then Claude deserves a serious trial. If your team needs image creation or voice or one general app for many small tasks then ChatGPT probably fits better. If AI is central to your operations there is a strong case for using both and letting each handle what it does best.
Whatever you choose run your own real prompts on both before committing. Use a real email thread or a real spreadsheet or a real document your team works with. Ten minutes of that test will tell you more than any comparison article including this one.
The Bottom Line
Claude and ChatGPT are both excellent and both are improving almost monthly. The gap that mattered a year ago has narrowed and the differences that remain are mostly about product focus rather than raw intelligence. Pick the one that matches your work. Then review the choice every few months because this space is not standing still.