is the one everyone's most excited about. The hype around this model is kind of insane. One poster is saying that it will show how ahead Open AI really is compared to the competition. That's a pretty bold claim, especially considering how competitive the space has gotten recently. We've seen Claude, Gemini, and other models really closing the gap. So, if GPT 5. 3 really does pull ahead by a significant margin, that would be huge news. Of course, OpenAI has been relatively quiet lately, while everyone else has been shipping updates and leaks. So, maybe they've been cooking something special behind the scenes. Then, we got Gemini 3 Pro and Gemini 3 Flash G. Google stepping up with not one but two models moving to general availability at the same time according to these reports. Now the pro version should bring more advanced capabilities, better reasoning, more complex problem solving, that kind of thing. And Flash is all about speed and efficiency, which is really important for production use cases where you need quick responses and lower costs. Google's been on a roll lately with their AI releases, and moving these to general availability means they're confident enough in the stability and performance to open them up to everyone. And finally, we got Gro 4. 2 or 420. Yes, that's the actual version number. Very on brand for Elon. XAI is bringing their own contender to the arena. Now, here's where things get really interesting and controversial. A user named Sharasula, who appears to have insider knowledge or early access to some of these models, made a bold claim that all the models I just mentioned will be defeated by something called Fenick. They're referring to what they claim is an upcoming Sonnet model. Cherula goes on to say that they haven't actually tested GPT 5. 3 personally yet, but they have high hopes for its math and law capabilities specifically. Those are areas where language models have been improving rapidly and apparently GPT 5. 3 might excel in that category. But here's the really interesting part. In their testing, Chedisoula claims that Fenk is one generation ahead of Snow Bunny, which was Gemini's leaked model. Now, if Fenk really is an upcoming Claude Sonnet model, and if it's truly a full generation ahead of what we assume is the current Sonnet, that would be absolutely massive for Enthropic. They have been competitive and are starting to lead the pack. Of course, and I can't stress this enough, we should take all the leaks and insider claims with a huge grain of salt until we see official announcements, verified benchmarks, and realworld testing from multiple independent sources. The AI community loves hype and speculation, and sometimes these predictions don't pan out. But the fact that someone with apparent inside knowledge is making these specific claims has definitely got people's attention. Let's talk money for a second because pricing is crucial in this space. One user in the thread asked about how much fenck would cost and cherula responded that it would be cheaper than Opus 4. 5 specifically half the price. Now that's really significant. If you can get better performance at half the cost, that's a gamecher for anyone building applications or using these models regularly. Cost has been one of the biggest barriers to adoption for most powerful models. So, if Enthropic can deliver top tier performance at a much lower price point, that could give them a serious competitive advantage. As for timing, when someone asked about the release dates, the response was max 2 weeks. So, we could literally be seeing these models drop any day now. Some might come next week, some the week after, this timing makes perfect sense when you think about where we are in the AI race. Every company is trying to establish dominance and grab market share. OpenAI, Google, XAI, and Anthropic are all making major moves simultaneously. It's like they're all trying to oneup each other, and we're the beneficiaries because we get access to increasingly powerful tools. The next two weeks are going to be absolutely wild for anyone following AI development. We're going to see real benchmarks, actual performance comparisons, hands-on testing from developers and researchers, and we'll finally find out if these leaks and predictions hold up. Are the performance claims accurate? Will the pricing be what's promised? Does Fenk really exist? And is it really that good? We're about to get all these answers very soon. If beating Snow Bunny, the leaked Gemini model, wasn't enough. Apparently, this new model, which is going to be called Claude Sonnet 5 most likely, is supposed to be cheaper and faster than Opus 4. 5
