rihim

incoming @EverpureData | computer engineering @UCSB class of 2027 | @ucsbNLP | prev: swe intern @Cisco

[rihim @rihim_s]

@baynkr you’re doing this wrong you have to make it start by making it figure out a way to make planck time smaller

[rihim @rihim_s]

@MIntellego @francoisfleuret yeah this is really fun

[rihim @rihim_s]

its cause its basically getting martyred

[Tyler @rezoundous]

Fable 5 is the first model I've seen that has near unanimous positive feedback on X

[rihim @rihim_s]

@abacaj not online but irl yeah

[rihim @rihim_s]

@ericzakariasson can i get testflight when it launches

[rihim @rihim_s]

@Hay155372 @xeophon yeah ik im saying they could fine tune/distill a version specifically for codebase exploration to be faster/cheaper

[rihim @rihim_s]

@HououinTyouma i’m getting the opposite now they say all my ideas are dead ends and suck and i should give up and there’s no point in continuing on it

[rihim @rihim_s]

@_ueaj @AndrewCurran_ that fully makes sense given how limited the interface is to use llms - just writing text strips so much nuance and taste that knowing how to prompt it really matters a lot. seems like fable had that intuition of understanding built in from the other side as well

[rihim @rihim_s]

@DWestkrew @thesnufki @haider1 not necessarily, it can outperform mythos AND they don’t have to compare it cause it isn’t an accessible model. kinda shady but i can see an argument being made that it’s the best frontier model available

[rihim @rihim_s]

@ThiccQuidity @Kalshi that this time the government banned it in less than a week after release

[rihim @rihim_s]

@_ueaj @AndrewCurran_ damn i didn’t use it out of principle cause of the ML research lora nerf but now i wish i had

[rihim @rihim_s]

@hopes_revenge

[rihim @rihim_s]

@giffmana

[rihim @rihim_s]

@teortaxesTex yeah 4.6 was quite special, surprised 4.7 is higher than 4.8; 4.7 seems more rl’d to me

[Shannon Sands @max_paperclips]

There's a much funnier thing ensembles unlock though (if it's consistent). It doesn't matter if it's inefficient, really. if you can get Mythos by throwing 3 or 4 other weaker models in a trenchcoat, you can distill from the ensemble directly. I wonder, how many Qwens + Kimi's + Deepseeks + GLMs do you need to throw in a trenchcoat to get Mythos quality data? Can you stack enough 9b's to reach heaven?

[rihim @rihim_s]

@max_paperclips @LokiJulianus @teortaxesTex the only issue is the model providers themselves limiting it lmao; ironically neither of the actual model developers can do it cause they can’t use their model in that way

[rihim @rihim_s]

@gfodor speak softly and carry a big stick

[rihim @rihim_s]

@viemccoy i'd assume the main issue is cost, right? given how inefficient some of the current models are with token output it'd probably cost a multitude more than running the good model (ofc it's extraordinary times rn)

[rihim @rihim_s]

@mweinbach slightly unrelated but the new ui looks so damn good here, i thought it was a physical glass thing oval at the beginning of the video