361–370 of 527 posts
Earlier quoted context omitted.
Long term memory in an LLM is its weights.
Not really, because humans can form long term memories from conversations, but LLM users aren’t finetuning models after every chat so the model remembers.
Earlier quoted context omitted.
Long term memory in an LLM is its weights.
Not really, because humans can form long term memories from conversations, but LLM users aren’t finetuning models after every chat so the model remembers.
Users can do that if they want, but it’s more effective and more efficient to do that after every billion chats, and I’m sure OpenAI does it.
Earlier quoted context omitted.
I had a philosophical discussion with o3 model earlier today. It was much better than 2.5 pro. In fact it was pretty much what I would expect from a professional philosopher.
I'm not expecting someone paying $200 a month to access something to be objective about that particular something. Also “what I would expect from a professional philosopher”, is that your argument, really?
What’s wrong with my argument? You questioned the performance of the model on non-STEM tasks, and I gave you my impression.
Very impressive! But under arguably the most important benchmark -- SWE-bench verified for real-world coding tasks -- Claude 3.7 still remains the champion.[1] Incredible how resilient Claude models have been for best-in-coding class. [1] But by only about 1%, and inclusive of Claude's "custom scaffold" augmentation (which in practice I assume almost no one uses?). The new OpenAI models might still be effectively bes…
The image generation improvement with o4-mini is incredible. Testing it out today, this is a step change in editing specificity even from the ChatGPT 4o LLM image integration just a few weeks ago (which was already a step change). I'm able to ask for surgical edits, and they are done correctly. There isn't a numerical benchmark for this that people seem to be tracking but this opens up production-ready image use case…
After refreshing the browser I see that the old o3-mini-high has gone now so I continued my coding task conversation with o4-mini-high. In two separate conversations it butchered things in a way that I never saw o3-mini-high do. In one case it rewrote working code without reason, breaking it, in the other it took a function I asked it to apply a code fix to and it instead refactored it with a different and unrelated…
Earlier quoted context omitted.
The image generation improvement with o4-mini is incredible. Testing it out today, this is a step change in editing specificity even from the ChatGPT 4o LLM image integration just a few weeks ago (which was already a step change). I'm able to ask for surgical edits, and they are done correctly. There isn't a numerical benchmark for this that people seem to be tracking but this opens up production-ready image use case…
wait, o4-mini outputs images? What I thought I saw was the ability to do a tool call to zoom in on an image. Are you sure that's not 4o?
I have a very basic / stupid "Turing test" which is just to write a base 62 converter in C#. I would think this exact thing would be in github somewhere (thus in the weights) but has always failed for me in the past (non-scientific / didn't try every single model). Using o4-mini-high, it actually did produce a working implementation after a bit of prompting. So yeah, today, this test passed which is cool.
Unless I'm misunderstanding what you are asking the model to do, Gemini 2.5 pro just passed this easily. https://g.co/gemini/share/e2876d310914
But generally, o1-pro listens to my profile instructions WAY better, and it seems to be better at actually solving problems the first time. More reliable.
But they are all quite similar and so far these new models are similar but faster IMO.
On one hand the answers became a lot more comprehensive and deep. It’s now able to give me very advanced explanations.
On the other hand, it started overloading the answers with information. Entire concepts became single sentence summaries. Complex topics and theorems became acronyms. In a way I’m feeling overwhelmed by the information it’s now throwing at me. I can’t tell if it’s actually smarter or just too complicated for me to understand.
It's pretty frustrating to see a press release with "Try on ChatGPT" and then not see the models available even though I'm paying them $200/mo.
Earlier quoted context omitted.
"Who was the President of the United States when Neil Armstrong walked on the moon?" Gemini 2.5 refuses to answer this because it is too political.
I call bs on this: https://g.co/gemini/share/ed38e9d38b02