Comment by observationist
9 hours ago
You can make a good guess, though - OpenAI had a significant lead, its moat was being a generation or 2 ahead of the competition, and as of the end of 2025, OpenAI, Google, xAI, and Anthropic are pretty much neck and neck. Grok and Gemini are likely to be the top 2 within the next couple months, and the Chinese open models are hot on their heels.
OpenAI is going to be competing for third or fourth place with Anthropic unless one of them pulls off a big capabilities or efficiency leap while remaining believably as good as the other top models. Google and xAI have advantages that the others don't, and are capitalizing on them like crazy. It remains to be seen whether xAI can compete with the Google hardware advantage, but they have economies of scale, differences in mission, and Elon's billions on their side, so it's turning out to be a very interesting race.
Sama could also finagle a funding rabbit or strategic partnership out of his hat and also have the next top tier model, amazing everyone again and keeping a plausible "best in class" lead for a while; OpenAI would have to be down for at least a year before I counted it out completely. It's not looking very pretty for them right now, though.
No one believes Grok will be top 2 in a couple months. OpenAI and Gemini, in one of the two orderings, will continue to be far ahead of Grok in "the next couple of months". I encourage you to bookmark your claim here and return in 2 months to take stock in your ability to predict/bluff.
How do you know which two models are 'top 2' and how do you expect the parent commenter to know which are top 2 when he comes back to this comment and verify his prediction?
Genuine question.
Grok even if it were technically superior is hampered by it being owned and run by a mercurial racist narcist who in drug infused mania keeps accidentally turning it into Mecha-Hitler in attempts to make it fit his political views or turning it into a generator of child porn. No one in their right mind wants to have their product associated with that kind of liability. Who wants an ai agent that starts spewing great replacement theory at their customers?
Even if OpenAI gets “better” than Google, it’s still a sharecropper on other people’s infrastructure and depending on more expensive Nvidia chips while Google has a real business that throws off 10s of billions a year in profit, it’s own infrastructure, its own TPUs, and actual customers with both search and related and GCP.
The open source models are showcasing a lot of cool research ideas that will undoubtedly end up in future frontier models from OpenAI and others. We're at a point where the big proprietary AI firms barely do any foundational research anymore, they're purely about incremental improvement. All the proven substantial progress recently has been coming from the open models side.
What I don't understand is why Gemini is not #1, other than that Google has no economic reason to have the same fire under its ass to be #1 as Anthropic and OpenAI. Or maybe they are correctly assessing that getting to good enough and out-building infrastructure is more valuable; they do have their TPUs as bets on their future and their search monopoly today to print nearly endless free cash flow. Perhaps Gemini is advancing at exactly the right rate for them.
I guess there is one thing that Gemini is objectively better at than either, which is long context, and it does seem to be by an order of magnitude. What boggles my mind is why Gemini is still not as good as the open weight frontier models yet. If they got just to parity with that along with their existing long context and strong token pricing, they'd be able to take over the coding market. Are they just biding their time to make their move? Hard to discern.