[dead]
GLM-5.1: Towards Long-Horizon Tasks
71–80 of 285 posts
Re: GLM-5.1: Towards Long-Horizon Tasks
#72Earlier quoted context omitted.
I'm on their pro plan and I respectfully disagree - it's genuinely excellent with GLM 5.1 so long as you remember to /compact once it hits around 100k tokens. At that point it's pretty much broken and entirely unusable, but if you keep context under about 100k it's genuinely on par with Opus for me, and in some ways it's arguably better.
100k tokens it's basically nothing these days. Claude Opus 4.6M with 1M context windows is just a different ball game
Claude Opus can use a 1M context window but I’ve found it to degrade significantly past 250k in practice.
Re: GLM-5.1: Towards Long-Horizon Tasks
#73Comments here seem to be talking like they've used this model for longer than a few hours -- is this true, or are y'all just sharing your initial thoughts?
My local tennis court's reservation website was broken and I couldn't cancel a reservation, and I asked GLM-5.1 if it can figure out the API. Five minutes later, I check and it had found a /cancel.php URL that accepted an ID but the ID wasn't exposed anywhere, so it found and was exploiting a blind SQL injection vulnerability to find my reservation ID. Overeager, but I was really really impressed.
This is insane, I love it.
Re: GLM-5.1: Towards Long-Horizon Tasks
#74Earlier quoted context omitted.
Is manual compation absolutely mandatory ?
I haven't screenshotted to alas, but it goes from being a perfectly reasonable chatty LLM, to suddenly spewing words and nonsense characters around this threshold, at least for me as a z.ai pro (mid tier) user. For around a month the limit seemed to be a little over 60k! I was despondent!! What's worse is that when it launched it was stable across the context window. My (wild) guess is that the model is stable but z.…
I have gone back to having it create a todo.md file and break it into very small tasks. Then i just loop over each task with a clear context, and it works fine. a design.md or similar also helps, but most of the time i just have that all in a README.md file. I was also suspicious around the 100k almost to the token for it to start doing loops etc.
Re: GLM-5.1: Towards Long-Horizon Tasks
#75Comments here seem to be talking like they've used this model for longer than a few hours -- is this true, or are y'all just sharing your initial thoughts?
My local tennis court's reservation website was broken and I couldn't cancel a reservation, and I asked GLM-5.1 if it can figure out the API. Five minutes later, I check and it had found a /cancel.php URL that accepted an ID but the ID wasn't exposed anywhere, so it found and was exploiting a blind SQL injection vulnerability to find my reservation ID. Overeager, but I was really really impressed.
That is both amazing and terrifying.
Re: GLM-5.1: Towards Long-Horizon Tasks
#76To be honest I am a bit sad as, glm5.1 is producing mich better typescript than opus or codex imo, but no matter what it does sometimes go into shizo mode at some point over longer contexts. Not always tho I have had multiple session go over 200k and be fine.
Isn't the same with opus nowadays?
Re: GLM-5.1: Towards Long-Horizon Tasks
#77To be honest I am a bit sad as, glm5.1 is producing mich better typescript than opus or codex imo, but no matter what it does sometimes go into shizo mode at some point over longer contexts. Not always tho I have had multiple session go over 200k and be fine.
I honestly still hold onto habits from earlier days of Claude & Codex usage and tend to wipe / compact my context frequently. I don't trust the era of big giant contexts, frankly, even on the frontier models.
I also feel like its helping me on the big models these days with claude giving so many issues.
Re: GLM-5.1: Towards Long-Horizon Tasks
#78Earlier quoted context omitted.
Is manual compation absolutely mandatory ?
I haven't screenshotted to alas, but it goes from being a perfectly reasonable chatty LLM, to suddenly spewing words and nonsense characters around this threshold, at least for me as a z.ai pro (mid tier) user. For around a month the limit seemed to be a little over 60k! I was despondent!! What's worse is that when it launched it was stable across the context window. My (wild) guess is that the model is stable but z.…
basically my expirience as well. Sometimes it can break past 100k and be ok, but mostly it breaks down.
Re: GLM-5.1: Towards Long-Horizon Tasks
#79Comments here seem to be talking like they've used this model for longer than a few hours -- is this true, or are y'all just sharing your initial thoughts?
My local tennis court's reservation website was broken and I couldn't cancel a reservation, and I asked GLM-5.1 if it can figure out the API. Five minutes later, I check and it had found a /cancel.php URL that accepted an ID but the ID wasn't exposed anywhere, so it found and was exploiting a blind SQL injection vulnerability to find my reservation ID. Overeager, but I was really really impressed.
Yeah it seems they did not align it to much, at least for now. Yesterday it helped me bypass the bot detection on a local marketplace. that i wanted to scrap some listing for my personal alerting system. Al the others failed but glm5.1 found a set of parameters and tweaks how to make my browser in container not be detected.
Re: GLM-5.1: Towards Long-Horizon Tasks
#80Earlier quoted context omitted.
> It's an okay model. My biggest issue using GLM 5.1 in OpenCode is that it loses coherency over longer contexts Since the entire purpose, focus and motivation of this model seems to have been "coherency over longer contexts", doesn't that issue makes it not an OK model? It's bad at the thing it's supposed to be good at, no?
long(er) contexts (than the previous model) It does devolve into gibberish at long context (~120k+ tokens by my estimation but I haven't properly measured), but this is still by far the best bang-for-buck value model I have used for coding. It's a fine model
i have glm and kimi. kimi was in most of the cases better and my replacement for claude when i run out of tokens. Now im finding myself using glm more then kimi. Its funny that glm vs kimi, is like codex vs claude. Where glm and codex are better for backend and kimi and claude more for frontend.
as kimi did a huge amount of claude distilation it seems to be somewhat based in data
https://www.anthropic.com/news/detecting-and-preventing-dist...