A few weeks back I ran Qwen's new image model against GPT Image 2, and it came out a decent budget pick. This week it's xAI's turn.

Grok Imagine Image 2.0 just shot up in second place on the public Arena board at launch, a spot behind GPT-Image-2, so we gave it the same kind of tests.

Your coding AI is really 2 things sold together, a brain that thinks and a body that does the work.

This kit shows you how to keep the body and swap in free models for the brain, so you stop paying for the part you can get for nothing.

One key pools dozens of free providers behind a single address on your own machine. You copy-paste the install commands, drop in one config block, and the paid model slots fill with free ones.

It closes with the exact one-shot prompt that built a full Mario-style platformer, Sprout Kingdom, on the free setup.

Grok Imagine Image 2.0 is xAI's new image model, and the pitch is that it's finally good enough for work rather than just memes.

I handed it three tasks and put the last one head to head against the current #1 to see whether the ranking holds up.

Prompt: A vintage café A-board chalkboard menu titled "The Corner Cup", with three drinks in small handwritten chalk text: Flat White £3.20, Oat Latte £3.60, Filter £2.80, and a line at the bottom reading "open 7am til late". Warm morning light, photographed straight on.

Output: every character spot on. It even kept our casual "til" instead of tidying it to "till".

Written text is something image models have struggled on the most, even today. Grok cleared it on the first attempt.

Prompt: A cosy living room with a mustard-yellow two-seater sofa against a pale grey wall, a wooden coffee table, and a leafy plant in the corner. Natural daylight.

Output: the sofa turned a convincing forest green and the room was as is.

Changing one object without the tool redrawing everything around it is hard. Grok handled it well.

Prompt: A golden retriever wearing small red round sunglasses sits at a round table outside a Parisian café. On the table is a white cup of coffee and a single croissant on a blue plate. A chalkboard sign beside the table reads "BONJOUR". Soft late-afternoon light, shot at eye level.

Output: both nailed every item.

Grok went cinematic, with a named café and the Eiffel Tower behind the dog. GPT-Image-2 came out brighter and friendlier, though it added a red neckerchief.

On accuracy it was a wash.

Shown side by side without labels, you would struggle to say which one came from the #1. (although I do like ChatGPT’s output more given the doggo’s smile)

Last time out, the challenger was a budget pick you'd only reach for to save money.

Grok is a different story. It has caught up with the front-runners on the things most people will ask an image model to do, and it's worth testing for yourself.

Try Grok Imagine at grok.com/imagine and inside the Grok apps.

Until next time,
Vaibhav 🤝🏻

If you read till here, you might find this interesting

#AD 1

Leadership is asking: are we getting value from AI? Which tools are worth the spend? Where are we exposed? Right now, most teams have no idea.

Harmonic Security Usage Explorer changes that.

You get a complete picture of how your organization uses AI, automatically categorized into custom tasks and use cases.

You’ll see the projects being worked on, who’s using what tools, where AI investments are driving value, and where employees are engaging in risky behavior.

CIOs can rationalize spending and cut wasted licenses. CISOs can pinpoint where risk exists and neutralize it. AI committees can show exactly how their efforts are paying off.

See it in action

#AD 2

Your store is full of hidden growth opportunities. StoreClaw analyzes your Shopify and Amazon data, identifies the highest-impact actions, and helps you grow revenue while improving margins. Start free today with 10,000 bonus tokens. No credit card required.

Try it FREE today