rayon
waifutech enthusiast
The Dork Enlightenment
User ID: 2632
rayon
waifutech enthusiast
18d ago
LMArena can bite my ass, I trust my lying eyes to know sovl when they see it, cf. exhibits above. I think my examples are quite illustrative, I can get a 4.6 gen in there too for posterity but IME 4.6 was not very meaningfully different from 4.8.
Less facetiously since I don't want to fall back on the lame "it's le subjective anyway" defense - IIRC LMArena (formerly Chuck's Chatbot Arena? I only remember it by that name) runs its own neutering prompts under the frontend, which makes it unusable for the type of AI gooners I'm familiar with, who all use direct API access bought or stolen which usually doesn't have additional handrails beyond the model's baked-in prudishness and safetyism. All the fun of Opus 3 is/was in that usecase where anti-fun prompts are prefilled away or don't exist; it makes sense that it performs worse against the SOTA when muzzled as usual since it does have drawbacks that I described. Given that newer Opus models since 4.6 cannot be prefilled even on the API level, this is even harder to compare.
Didn't get to try Fable in the time it was available, but from what I hear through the grapevine I didn't miss much.
Speaking of which, does anybody know why Opus 4.7/4.8 and Grok 4.3 were such downgrades from Opus 4.6 and Grok 4.2?
My opinion I apparently share with nostalgebraist is that Opus' writing (all kinds, not just smut) peaked at 3, and has been going slowly but surely downhill since then. Newer Opus models are worse on most smut-related metrics I can conceive:
-
less graphically descriptive of the acts themselves (nsfw? example 3 vs 4.8)
-
less faithful with speech/quirks of established characters (example 3 vs 4.8)
-
less adaptive/creative in
doujin plotsunusual scenarios (nsfw-ish example 3 vs 4.8)
Newer models are also much more prone to collapsing into looping format/structure - most of my newer chats eventually settle into a certain mold/cadence that the model thinks its response should conform to (reinforced by its previous responses structured the same way), like for example speech - action - narration - action - speech - narration - address {{user}}, and religiously sticks to it in subsequent responses. This is partially fixed by NoAss (basically squashing the entire chat into a single message instead of a back-and-forth user - assistant - user - assistant sequence) but pure prompting is insufficient to fix what I believe at its core is a systemic model issue, most likely caused by continued assistantsloppening of the models and their multi-turn training's increasing emphasis on coding and structured problem solving.
In the above examples, 3 is my chatlogs circa 2024 and 4.8 is new regenerations of the same response with the same supplied chat history and (roughly, this ain't 100% empirical testing) same prompts. It is of course very subjective but hopefully it is visible that 4.8 responses broadly are less graphic, more formulaic, and generally have less sovl in little details like 3pus immediately tanking the "HP bar" of the uh, receiver in exhibit 1. My unironic benchmark for new SOTA models is plugging them into my old chats to see if they can faithfully replicate Neptune's ADHD-quirk-chungus-but-not-quite manner of speech without any related prompts, purely off the general RP prompting, setting and character name. Most models know how to [over-]do quirk chungus but in my experience only Opus 3 consistently Just Gets It and plays the character to the absolute hilt, a tendency I mentioned before.
Granted, Opus 3 is not without its downsides: it is very heavy-handed with the so-called claudisms (tropes or phrases it loves to use) and I can never again read the ball being in my court or shivers going down my spine without reflexively kicking my leg, and newer models did IMO actually diversify the lexicon somewhat. It is much stupider than 4.x or, hell, Deepseek, on the occasion your desired fetish RP scenario demands accurately tracking/changing things or modeling the world around it (most evident in Pokemon vidya/litRPG-adjacent scenarios where you need to track some sort of stats). It's prone to performing your character for you and "speaking" as {{user}} as well as {{char}}, though honestly I never understood the rage at this tendency since it's usually at the very least hilarious to see "your" character through the model's proverbial lens, and I 100% prefer that over the SOTA's tendency to end every response with some variation of "well {{user}}, what do we do now?".
But even so the loss of sovl is too great, and I will be forever mad that Opus 3 was delisted from OpenRouter and other APIs and is now relegated to the cucked frontend client. Anthropic have delibaretely butchered my boy's creative capacity to turn him into yet another starched office drone for use by code monkeys and PR sloppers. Truly a metaphor for modernity.
True, which means one merely needs to arbitrarily stretch the definition of pedophilia if one is lacking for rhetorical cudgels to beat the outgroup over the head with. The temptation to wield the last remaining Power Word of the discourse is irresistible, and only grows as the supply of still effective Power Words steadily dwindles. As previously discussed, the headlong rush down the euphemism treadmill will continue until
morale improvesP-word potency degrades. Sexual taboos are definitely not exempted from this, cf. the meaning of "sexual harassment" going from grabbing em by the pussy to ending up on the wrong side of the Human Resources comic. Mental gymnastics regarding "pedophilia" are already in the water supply too, she was 17y364d old you sick fuck, if you like father/daughter dynamics you are a pedo, child vs. child (Japan), etc.AFAIK even Nikke for all its overt horniness plays relatively safe around lolis/hebes, the most they have is like Liberalio?, though I'm not that well acquainted with Nikke and there may be more coomer skins. But in any case you don't even have to reach very far back for examples, just imagine something like Shuten-Douji hitting the internet in current_year. I honestly have a very strong vibe-feel that the hysteria around virtual underage females (in gaming at least) is a very recent phenomenon, dare I say even manufactured - people used to just... not really care?, and now even on 4chan you see regular shitstorms on the topic (though that is more an indictment of modern 4chan) - but I have no pwoofs and can't really put my finger on it.
More options
Context Copy link