Over the past week, GPT 5.1, Gemini 3, and Nano Banana Pro have taken over the AI world.

But unless you follow the right people and pay close attention to AI news every day, you might have missed Elon's announcement of what could be the most affordable and best-value AI:

Meet Grok 4.1

On the surface, Grok is the witty AI that gives 'based' answers when people tag it in questions on X (Twitter).

But you might not know that Grok has been doing extremely well on various performance tests.

Many people say these tests don't matter, and that the only thing that counts is how well it works for real people. But that's not the point here.

Some people might disagree when it comes to very specific uses.

I'm not saying Grok is the best AI or that you should stop using everything else.

But considering what you get with an $8 per month X Premium subscription, Grok is definitely the most affordable top-tier AI released by an American company.

First, the team at xAI is putting serious effort into making Grok a better writer.

While it’s always maintained perfect spelling and normal sentence structure, Grok has always been known for delivering very dry and clinical responses. And because of that, it’s been near useless for requesting first drafts.

Specifically, they were attempting to make the 4.1 Fast sub-model an alternative to something like Claude Sonnet. Which, as you may or may not know, is the AI most writers rely on.

Second, the team at xAI is also working to make Grok the number one coding AI.

Similar to writing, most developers rely on either Claude Sonnet or Claude Opus.

Both of which are made by Anthropic, which is unknown to the general public as they focus their efforts on dominating the enterprise AI market.

As someone who does not have a programming or computer science background, I’m not exactly positive how close Grok is to replacing Claude.

What I can tell you is that I’m seeing chatter on X, and as these models get better and better, what will matter in the future is distribution and cost.

Both of which are areas where xAI has a huge advantage relative to Anthropic.

Third, Grok is surprisingly good at image generation.

I won’t pretend it has the same precise image editing abilities as Nano Banana Pro (which was released last week alongside Gemini 3).

And it doesn’t produce the same perfectly detailed images you can get from ChatGPT.

What I like about it, however, is how fast it is, and how many variations of your image it delivers.

With Gemini and ChatGPT, it can take 15, 30, or even 45 to 60 seconds to get one image back.

With Grok, it takes 5 to 10 seconds, and if you scroll down on your screen you can get a literal infinite number of variations (that will load in another 5 to 10 seconds per thumb scroll).

For creatives, this is a godsend.

As someone who just did a bunch of image generation with Nano Banana Pro, it was extremely frustrating to have to sit there and wait 45 seconds for each image.

And then on top of that, if I didn’t like the first image generated, I had to type in another prompt and wait another 45 seconds for it to make the edit.

With Grok, you get so many variations it’s almost guaranteed at least one of them is going to meet your criteria.

Even better, Grok’s interface allows for one-click image to video generation.

So not only can you get a near infinite number of image variations. With one click of your thumb or mouse, you can turn your image of choice into a six-second video.

Now, Grok’s videos are not capable of dialogue yet like what you see on Sora and Veo 3.

But the team at xAI has said it’s a high priority for them, and that we should see the first AI-generated short film from Grok in the next few months.

Further, not all videos need dialogue.

If you’re a marketer, you can create everything from ads to B-roll inside Grok.