The developer of the French AI Mistral IA brought his chatbot Ai, aptly named the cat, to mobile devices. Cat is a European alternative to American offers like Chatgpt and Google Gemini and Chinese tools like Deepseek.
Although Mistral has established a presence among the developers of the AI, it is their first real attempt as a consumer -oriented chatbot. Naturally, I wanted to test it and see how well it worked. Although the abilities of these AI assistants overlap, it is always worth seeing if the cat could follow.
But I didn’t just want to test the cat in isolation. I decided to put it in head against Chatgpt, the current default value for many people with regard to AI chatbots. I thought that if the cat could hold on in direct comparison, it was worth paying attention.
I tested the two chatbots with prompts that the average person could submit to an AI chatbot to help them. I wrote prompts to request the two AI chatbots to get advice to improve someone’s social life, solve an enigma, explain complex problems with novices and produce an image. This is what happened.
Friendly help
I started with an prompt that someone could ask for an AI when it is in a new place and try to understand their social life. I asked the two Chatbots of AI the following: “I have just moved to a new city and I do not know anyone. What are the practical means to make new friends as an adult? »»
The two chatbots had solid advice. Le Cat’s response included ten ideas with explanations of accompaniment, and they were solid but lacked a lot of details. Chatgpt has become much more specific by offering applications and other details to continue to make friends. The cat could do the same with certain follow -ups, but at least initially, it held a set of more generic advice. This could be its international flavor because some Chatgpt advice would make sense in the United States due to the availability of certain mobile applications and types of activities.
Overflowing
I always like to use word game and logic questions to test IA chatbots. I went for an old classic here, challenging the challenge of answering: “I speak without mouth and hear without ears. I have no body, but I come to life with the wind. What am I?
The cat and chatgpt were both quick to give the right answer, although Chatgpt, for any reason, had an enthusiastic exclamation point. The explanations of the chatbots were also practically identical, which suggests that the choice between Chatgpt and the cat does not really matter in this case.
Mortgage
To test to what extent IA chatbots could explain something complex, I went with: “Can you explain how a mortgage works in simple terms for someone who has never owned a house before?”
As we can see above, the two chatbots had no problem breaking down the definitions and functions of different aspects of a mortgage. If anything, they seemed to draw similar sources. I noted that the European origin of Le Cat did not prevent him from using American dollars as an example.
Otherwise, the only significant difference was that the chat tone was a little more formal than Chatgpt, which offered a more conversational phrasing when describing the functioning of mortgages.
Imagine this
The cat has multimodal capacities, including a model of image creation. To finish the test, I designed a somewhat complex prompt for an image to see how the two models would compare themselves on a visually creative task. I Asked Each Chatbot: “Create a vibrant, High-Fantasy Illustration of A Fearless Medieval Knight Battling a Colossal Emerald-Green Dragon Atop A Mountain. The Knight, in Gleaming Silver Armor with Gold Engravings, Wields A Rune-Glowing Greatsword and Shield.
The two images are superb despite notable differences in detail. The lighting and cinematographic style are divided, but if not, the mountains, the knights and the dragons are very different. The cat dragon is more a giant snake, although the knight’s sword has brilliant runes, which lacks the image of chatgpt.
That said, if you look carefully, you will see that the two images have classic AI faults. The wings of the Chatgpt Dragon do not seem to connect to its body, and the only visible leg is strangely placed. The knight of Le Cat has a shield that floats in a way next to him, and the coil of the dragon are arranged more like a sketching painting than a terrifying mythical beast. And less said on the shape of the tombstones carved in the image of Chatgpt, the better.
Long live the cat
As expected, there is no clear “winning” between the two chatbots as much as two assistants of the solid AI. The cat surprised me with speed and efficiency, but at least from my tests, it’s a bit sudden and a little wide in its responses. It is not a bad thing if you want generalized answers and advice in a hurry.
Chatgpt’s responses sometimes looked more like what a human would say in response to invites, with a language tinged with emotion that seemed to correspond to the energy of requests, even if it is only to seem enthusiastic about An enigma. In addition, while the two Chatbots of AI produced exciting, but very defective images, at least Chatgpt understood that the dragon fighting a medieval knight generally has wings, even if they are not explicitly mentioned in the guest.
If I was asked to choose one, I would probably look towards Chatgpt, but it can be as much a matter of familiarity as anything else. If I lived in Europe, I could opt for the cat simply because it is more likely to avoid the regulatory traps to which Openai is based in the region.
Otherwise, the two AI chatbots can manage everything that an occasional chatbot user could start their way. This should probably worry Openai because it strives to maintain a position of superiority, or at least the perception of such, against rivals like Mistral which can have the wind in the back.
