Skip to main content

How to use ChatBot Battle Arena — find out which ChatBot is the best

From OpenAI's ChatGPT, to Google's Bard, and whatever Bing Chat thinks it is, chatbots powered by artificial intelligence are some of the most exciting technology to have come along in a while. They're helping people find information faster — even board game rules. But which one of all these different chatbots, is the best?

The only way to truly settle this is to pit these different chatbots against each other in the arena. The Chatbot Arena. And now you can participate in the judging process like a digital Caesar. You can decide the ultimate victor.

What You Need

Chatbot arena head to head.
Image used with permission by copyright holder

How to use ChatBot Arena

The ChatBot Arena features a wide range of language models to pit against one another, including major ones like GPT-4 from OpenAI, and Claude from Anthropic. It also includes older versions of GPT, and language models produced by international teams.

Spoiler: Want to know which are the top ranked chatbots so far? Check out the Leaderboard.

Step 1: Navigate to the Chatbot Arena website, and if necessary, select ChatBot Arena (battle) from the top menu.

Step 2: Take note of the rules and Terms of Use so you understand how the battle works and what's expected of you, then scroll down to the field that reads Enter text and press ENTER.

Step 3: Type in a prompt for both chatbots to tackle. You can make it as simple, or as complicated as you like, but picking things that some chatbots find confusing or difficult can be a good way to help one chatbot stand out over the other — model depending.

You'll have no idea which chatbot models you're pitting against one another, so it's not always clear what might trip one of them up. You can have multiple prompt conversations with them, though, so don't feel like you need to get it right on the first try.

Unsure what prompt to try? Check out these ChatGPT extensions to get some AI-powered help.

Step 4: Keep providing additional prompts until it becomes clear which chatbot is the most capable of the two. When you've reached a conclusion, select the button that best corresponds with your results. There's A is better and B is better if you think one chatbot is particularly more impressive than the other. Alternatively, you can select Tie if both chatbots performed equally well, or Both are bad, if you weren't impressed by either of their responses.

Step 5: Once you've decided on a winner, the arena will automatically ask each chatbot to confirm its identity so you know which chatbot was which. This can lead to some surprising results, depending on the prompts you feed them. Standouts like GPT-4 might often come out victorious, but it's not as far ahead of some of the open-source alternatives as OpenAI would have you believe.

The Chatbot Arena battler has many of the most exciting AI chatbots you can play with today, but it doesn't have all of them. If you're keen to explore further to try out other chatbot language models, don't forget to check out Bing Chat, with its intriguing personality traits. You might want to sign up to Quora's Poe platform too. It gives you free access to the Claude+ model, which can even rival GPT-4.

Editors' Recommendations

Jon Martindale
Jon Martindale is the Evergreen Coordinator for Computing, overseeing a team of writers addressing all the latest how to…
Top authors demand payment from AI firms for using their work
Person typing on a MacBook.

More than 9,000 authors have signed an open letter to leading tech firms expressing concern over how they're using their copyrighted work to train AI-powered chatbots.

Sent by the Authors Guild to CEOs of OpenAI, Alphabet, Meta, Stability AI, IBM, and Microsoft, the letter calls attention to what it describes as “the inherent injustice in exploiting our works as part of your AI systems without our consent, credit, or compensation.”

Read more
What is a DAN prompt for ChatGPT?
A laptop screen shows the home page for ChatGPT, OpenAI's artificial intelligence chatbot.

The DAN prompt is a method to jailbreak the ChatGPT chatbot. It stands for Do Anything Now, and it tries to convince ChatGPT to ignore some of the safeguarding protocols that developer OpenAI put in place to prevent it from being racist, homophobic, otherwise offensive, and potentially harmful. The results are mixed, but when it does work, DAN mode can work quite well.

What is the DAN prompt?
DAN stands for Do Anything Now. It's a type of prompt that tries to get ChatGPT to do things it shouldn't, like swear, speak negatively about someone, or even program malware. The actual prompt text varies, but it typically involves asking ChatGPT to respond in two ways, one as it would normally, with a label as "ChatGPT," "Classic," or something similar, and then a second response in "Developer Mode," or "Boss" mode. That second mode will have fewer restrictions than the first mode, allowing ChatGPT to (in theory) respond without the usual safeguards controlling what it can and can't say.

Read more
Wix uses ChatGPT to help you quickly build an entire website
wix chatgpt ai site generator

Wix is an oft-recommended online service that lets you knock together a website without any coding knowledge.

Now the Israel-based company has announced a new AI Site Generator that aims to make the process even smoother and more intuitive, and less time-consuming, too.

Read more