Why Aren't We Talking More About GPT-4.1?
An AI model quickly gaining traction among software engineers who use Cursor/Cline/Roo/Kilo Code.
There's surprisingly little coverage about OpenAI's new API-only model, GPT-4.1, considering how it's quickly becoming one of the go-to AI models for coding.
Usage Consistently Tops the Charts
GPT-4.1 is overwhelmingly the most popular AI coding model in our free tier. It's so popular that we've landed on OpenRouter's leaderboard because of it.
Theo, a popular coding YouTuber, recently reported that GPT-4.1 is the third most popular AI model on his T3 Chat platform. Here are T3’s usage stats:
OpenAI also consistently tops the chart in the 'programming' category on OpenRouter, a chart which is considered a solid proxy for overall model popularity.
Positive Sentiment Is Growing
I've noticed overwhelmingly positive sentiment around GPT-4.1 in the AI coding community lately:
(source)
Here’s another one:
(source)
And they're not alone. There are dozens of similar tweets saying the same thing over the past few days.
Things aren't any different on Reddit, where this positive sentiment started the moment GPT-4.1 was released.
(source)
Still, this AI model hasn't gotten as much media attention as Claude and Gemini, and I wonder why.
Do We Love to Hate OpenAI?
If you've ever used AI to help you code, Claude was probably your top choice. Claude still remains the undisputed model of choice for coding.
Then there's Google. Gemini recently released 2.5 Pro and (as you can see from Theo's chart above) and it consistently ranks high, especially for AI coding:
Source: OpenRouter
OpenAI? Well, we're seeing more usage than hype (with the other model) to say the least. That's what inspired this blog post in the first place. GPT-4.1 deserves more coverage considering the usage stats (both on Kilo Code and other platforms).
As a software engineer, I’ve been taught to love hating on OpenAI in a way. Just take a look at the most popular stories whenever you search for OpenAI on Hacker News:
Not exactly glowing coverage. We've come to distrust OpenAI for a reason.
But that doesn't mean we shouldn't give credit where credit is due. GPT-4.1 seems to deserve that credit for bringing us closer to a future where we'll be more productive as software engineers.