@claudeai “we released a model which is barely better for 10x the cost. Openai will soon release 5.6 which will be 10x cheaper and on the same level or better. In short we once again generated slop.”
@DiscussingFilm I can literally smell the Wii resort map for some reason. I played this shit so much as a kid. Just aimlessly biking around the island with my footstep detection pad or whatever its called
@daniel_mac8 "twice as good". Literally anyone that has used both models know that 5.5 beats 4.8 by a landslide. Its not even close. this benchmark is shit and claude users are using it to cope. Deepswe is the only actual benchmark that shows the reality.
5.6 will blow mythos out
@ukhomeoffice how about you actually do something that has a positive impact on people. Like reducing living cost and improving living. No? didn't think so.
@cognition Not trusting this benchmark for shit. I have used both 4.8 and 5.5. 5.5 is faster cheaper and far better in every single instance. Deepswe still the best benchmark. anything with a claude model on top is slop.
@stupidtechtakes The more I see shit like this. the more I love open source. like why tf are you even in my business like this. You are AI platform. your only job is to provide me with llms that I pay for and tools.
Otherwise you can go fuck off.
@Star_Knight12 Well artificially made biological intelligence is in the works. Although still in the early stages, crazy things have been done.
Routing intelligence through an actual biological organism rather than through a machine would actually grant real General Intelligence.