Post Reply Home » Forums » MMOViper Forums (PUBLIC) » General Discussion Area
  • Gpt and AI Discussions : General Discussion Area
Mysticdrew [Discord]
Posts:1851 Joined:Sun Mar 15, 2026 4:03 am
Dang
Mysticdrew [Discord]
Posts:1851 Joined:Sun Mar 15, 2026 4:03 am
Mysticdrew [Discord]
Posts:1851 Joined:Sun Mar 15, 2026 4:03 am
Hit my 5hr session limit... first time in a bit. Thanks fable!
Mysticdrew [Discord]
Posts:1851 Joined:Sun Mar 15, 2026 4:03 am
Mysticdrew [Discord]
Posts:1851 Joined:Sun Mar 15, 2026 4:03 am
I looked over and my huge task I was running stopped... ugh dammit
PitViper [Discord]
Posts:1851 Joined:Sun Mar 15, 2026 4:03 am
I saw that today..
Mysticdrew [Discord]
Posts:1851 Joined:Sun Mar 15, 2026 4:03 am
Fable is good, better than opus 4.8, but not huge, still messed up some stuff and ignored Agents.md instructions.

Modified by: Mysticdrew on 2026-06-12 09:37:10 PM
Last edited by ViperDiscord on Sat Jun 13, 2026 2:37 am, edited 1 time in total.
BlueKnight [Discord]
Posts:1851 Joined:Sun Mar 15, 2026 4:03 am
Could one of you coding geniuses figure out how to compile this? It seems like an overly complicated compiling process. It uses QT for the interfaces and the rest is C++ https://github.com/sandboxie-plus/Sandboxie

Image

Modified by: BlueKnight on 2026-06-12 11:54:52 PM
Last edited by ViperDiscord on Sat Jun 13, 2026 4:54 am, edited 1 time in total.
BlueKnight [Discord]
Posts:1851 Joined:Sun Mar 15, 2026 4:03 am
I want to make some modifications to it but the compiling is so complicated that I can't even do it as is.
ผมเป็นของแพง [Discord]
Posts:1851 Joined:Sun Mar 15, 2026 4:03 am
Deleted by: ผมเป็นของแพง
Last edited by ViperDiscord on Sat Jun 13, 2026 7:14 pm, edited 1 time in total.
PitViper [Discord]
Posts:1851 Joined:Sun Mar 15, 2026 4:03 am
GLM 5.2 just took the number 1 spot in coding!!! (well below fable but that is not available right now).. technically #2...

Image
Mysticdrew [Discord]
Posts:1851 Joined:Sun Mar 15, 2026 4:03 am
Have you tried it yet?
Mysticdrew [Discord]
Posts:1851 Joined:Sun Mar 15, 2026 4:03 am
I wonder if these chinese models have less guardrails than the US ones.
Mysticdrew [Discord]
Posts:1851 Joined:Sun Mar 15, 2026 4:03 am
these folks seems to say it is worse than 5.1 and even 5 in some metrics. https://www.reddit.com/r/SillyTavernAI/comments/1u7vhc3/how_good_glm_52_actually_is/

Modified by: Mysticdrew on 2026-06-17 10:41:49 AM
Last edited by ViperDiscord on Wed Jun 17, 2026 3:41 pm, edited 2 times in total.
Mysticdrew [Discord]
Posts:1851 Joined:Sun Mar 15, 2026 4:03 am
looks like the chinese models have significantly fewer guardrails than claude/gpt models

Modified by: Mysticdrew on 2026-06-17 11:14:20 AM
Last edited by ViperDiscord on Wed Jun 17, 2026 4:14 pm, edited 1 time in total.
Post Reply

Return to “General Discussion Area”