Politicians that take cheap blows at someone like Elon are just doing it for political points. Never have they attacked Warren Buffett, why? There is a difference between real politics and what Gavin projects. A real politician would make good arguments that people from both sides would agree. Fair, equal systems which encourage society not ones where they put the country at risk and send job overseas.
@ClimateWarrior7 lol. Many homeless people, if not the majority have mental illness. Maybe you should be advocating for better treatment of mental illness to help the homeless. It would be a win/ win as you are seeming to be suffer as well.
I have used 4.8 now for a few tasks. It might be smarter, but sometimes too smart. It’s overthinking everything and make my workflow harder. I say commit everything, it thinks and says something else change something (me) I better do something about it do you want me to revert. I can’t trust it enough. I have gone back to 4.6 as it performs better.
Also I am making a workflow, so evertine I change sometime I trigger “workflow” surely you should have made it /workflow so you don’t keep reversing human language as your own commands
The problem with the ai generated one is that if the Batteries are on a skateboard underneath the car, then it cannot be this low. So the designers have a huge issue to distribute weight evenly around the car. This is something some car maker are doing well. The Luce tries to hide this issue, but it’s super ugly. Your ai concept is nice, but maybe can’t be made easily without people lying down
@Dominos_AU interesting issue I had today where I ordered on my phone and paid using apple pay. When I got to the store my order was attached to another users name. This brings up huge security concerns around my data and privacy. It seams you might tokenise payment data, but user data and details look to be less protected and possibly something that can lead to data breaches. At the very least it’s some edge case which associates payments against the wrong customers.
@tetsuoai What I don’t understand is why you are using grok when there are better llms for coding available. Surely having higher quality and more accurate is going to be significantly better?
Congrats @sama on 5.5, just upgraded after trying it. Changing it to my main driver over opus. Over the last 10m tokens it has performed better than opus in CC. Fixed issues that I was having trouble with in CC and seems to cover itself with better testing
@Wodeshed lol, says someone that hadn’t driven a Tesla. I drove my Porsche down the coast this week, I thought to myself on the way back that my S would have been easier and more pleasurable. I’ve driven it many times over 1000km with a few stops to charge
@mattshumer_ 4.7 will oneshot well. The issue is on large codebases it goes offline the rails, doesn’t follow plans and refactors code which it should not be touching. It will also finish early and not finish alll tasks until prompted.
@AnthropicAI I cannot believe how 4.7 does so poorly every time. I switched to 4.6 no problems for 6 hours. Opened another window, forgot to change to 4.6, so 4.7 went off the rails immediately. 20x more code and the wrong direction. Crazy @bcherny