Three people debate upcoming AI releases from OpenAI, Google, and Anthropic in the next two weeks. They're frustrated that model 4.7 feels worse than 4.6, especially at instruction-following, and blame compute constraints and heavy user traffic. They're hoping new releases actually fix things instead of repeating the same problems.
Is OpenAI oh so back at this point? You think they're gonna take off on a rocket ship? I don't know, man.
New release in 2 weeks. New release in 2 weeks. Everyone's releasing in 2 weeks. Google, OpenAI, maybe Anthropic. I don't know about Anthropic, but Google and OpenAI definitely releasing in 2 weeks.
What are they releasing? New pre-trains? More RL? Same base model?
More everything. More continued pre-training because the SpUD, they kind of like didn't finish the pre-training and just released it. So finish the pre-training. Do more RL, drop model. Google mostly just gonna be like multimodal slop. We'll see. I guess like my question is like, the narrative is that for even the normies who don't use fast mode, that 4.7 is worse than 4.6. Do you guys agree with this or no?
I think the instruction following has gotten objectively worse. I think they, it keeps missing Claude MD instructions or you pull a skill and you're like, dude, you didn't do exactly what was laid out in the skill. That seems to be a consistent Problem. I don't know. I feel like it's a compute problem more than anything else. It still has the like we've done a lot for today and go enjoy your weekend. I was like it's fucking Monday, like get back to work. It annoys me so much. It tries to like enforce me to stop working and I'm like clearly usage issue. But I think I think the four six golden age when it wasn't quantized in the beginning, that's when it was. Those were the days. Okay, fast mode four six pre pre the nerf. Oh. But I just think that there's this, this maturity of the models as they become more inference optimized and more people use them, that it becomes a worse experience at 10x or 100x the users. And I think that that's happened. Yeah, I just think 4.7 is fine. I think 4.6, 4.7 is probably the same. It's just that it's— oh my God, I saw that. It's kind of the same level of experience. Yeah, it's just too many users, bro. No, I need a Nimby. I need a Nimby AI. I need a Nimby Frontier model. Where no one else uses it except for, you know, except for me. So I can use it, I can get it at a higher, a higher rate. I mean, that's the appeal of Codex right now, I think, is that in theory you should be able to have higher rates, but— Can we talk about
