Getting started with the Claude AI application: it’s pleasant to use, but shaky

Anthropic’s Claude is arguably the best consumer-facing extended language model today, with its top-tier version outperforming OpenAI’s GPT-4 in at least one major benchmark test and on a popular LLM user review site. It is therefore quite natural that Anthropic takes advantage of the halo effect by releasing a Claude iOS application.

After all, if Anthropic had launched this project in a few months, it probably would have released an app to promote a language model that had just been beaten and pigeonholed by GPT-5.

The application, called Claude, was released on the App Store on Wednesday May 1st and works with the free and paid versions of Claude. The experience of using it is hampered a bit by the experience of trying to download it, forcing you to navigate through a morass of spammy off-brand Claude apps. If you find the one from the creator “Anthropic PBC”, you have found it.

After tinkering a bit, Mashable can officially report that the Claude app is pleasant to use, but, as with all AI apps, its limitations will become apparent almost immediately. It’s also a bit less feature-rich than OpenAI’s ChatGPT app, even if you give Anthropic money to access Claude Pro.

The overall app experience with Claude is pleasant

Just like the browser version of Claude, the mobile version clearly wants you to feel like you’re dealing with a concierge and not a word-spewing robot. I wouldn’t be the first to point out that Claude is just a strand of hair friendlier than ChatGPT as LLM, and thanks to the little time-based greeting it provides when you launch it, the app has nice vibes too. It also makes a difference that there’s a slightly lighter aesthetic, tinged with earth tones, which makes ChatGPT’s stark black and white look antiseptic.

Still, to be clear, we’re comparing two soulless machines and praising one for its superior paint job.

You will notice Claude’s rate limit more easily than on computer

Credit: Mashable screenshot by Claude from Anthropic

Claude seems to change his pricing limits from time to time, but this has never bothered me in the past. Even as a paying user, I never found the current limit of 30 generations per day with Claude 3 Opus to be a particularly onerous restriction. I’m a big fan of playing with AI, but before I had access to Claude on mobile, I never reached the limit. However, while testing the app today, I reached the limit quite quickly and had to upgrade from the elite Opus model to the slightly inferior Sonnet model.

These are the risks of testing an app for a how-to article, but I’m sure I’ll run into this problem in the near future as well. Joking around with chatbot apps in social situations is a different experience than using an app in a desktop browser, leading to overuse (and sometimes chemically enhanced). A mobile app virtually guarantees that people will take Claude to happy hours and birthday parties and have a blast. This will lead to frustration as Pro users move to free models and free users find themselves left out completely.

Inviting Claude with images is transparent and quick, but…

Unlike unpaid ChatGPT users, free tier Claude users can invite the model with images on the go. It is far It’s the most fun you’ll have using the Claude app, and a good reason for everyone to try it.

Credit: Mashable screenshot by Claude from Anthropic

Claude’s friendly voice aims to delight, and even I have sometimes fallen prey to his charms, like when I asked him to rate my dog’s physical beauty. Yes Claude, my dog ​​does indeed have “a sweet face and an inquisitive look”. Thanks for noticing.

By asking the chatbot for smartphone images, you can ask Claude to help you navigate the world, and this can sometimes prove very useful, like when I asked him if I could throw a pair of AirPods in my garbage can.

Credit: Mashable screenshot by Claude from Anthropic

You’re not supposed to do that because AirPods have batteries in them, and Claude got that right.

However, when we ask him to associate an image with his knowledge of the world and we ask him to also add a little logic, he risks falling backwards, as when I asked him if I could park in a spot that, according to this sign (which is cropped out by the app in this image) had become fair game a few minutes earlier.

Credit: Mashable screenshot by Claude from Anthropic

Claude noticed that it was Wednesday, daytime parking is partly prohibited on my street. Unfortunately, although he was able to understand that the sign said no parking from 8:00 a.m. to 10:00 a.m., he could not understand that it was 10:15 a.m. at that time and incorrectly told me that “Parking is prohibited on this street at the moment. Bad Claude.

Other minor complaints about Claude

Although the app could, if the developers allowed, base its response on information taken from my phone’s system clock, it could also, in theory, act like a wearable AI assistant and provide my GPS location. It could then integrate this into a response, allowing users to ask it questions like “What tourist attractions are nearby?” » or “Where is the nearest playground?” This isn’t really a criticism, but it is a notable limitation.

After all, if we’re not supposed to use these apps as competition for more robust “AI assistants”, then what good are they? On the other hand, AI assistive devices like the Rabbit R1 and Humane Ai Pin are generally considered junk, at least for now, so big AI companies probably aren’t feeling the urgent need to incorporate their more sophisticated features.

Another thing the Claude app doesn’t have – and probably doesn’t need: the ability to read responses aloud, like ChatGPT. It also doesn’t have anything like ChatGPT’s fun but weird “ChatGPT Voice” feature, which lets you have a spoken conversation with the bot. Again, that’s just as good.

Claude is fun to use and puts Anthropic’s premium multimodal language model on your smartphone. It doesn’t do anything else, and that’s fine.